Talk:L17-01: Difference between revisions

From Lojban
Jump to navigation Jump to search
(17 intermediate revisions by the same user not shown)
Line 122: Line 122:
.i lo stizu cu srana do -> "The chair is associated with you."
.i lo stizu cu srana do -> "The chair is associated with you."
.i lo stizu do co'e -> "The chair has an obvious/unimportant relationship with you."
.i lo stizu do co'e -> "The chair has an obvious/unimportant relationship with you."
 
==Scope
=== Issues: unsorted ones ===
*{{mun|mi na cadzu ki'u lo nu carvi|It's not that I walk when it's raining.}}
**Here, the speaker might imply that he or she likes going in the rain.
*a bridi creates a new scope
**but it inherits undeclared variables from parent scopes
*da scope.
*da scope.
**{da zo'u da broda .i je da brode} = {da zo'u ge da broda gi da brode} = {suzdza lo ka poi'i ge ke'a broda gi ke'a brode}
**{da zo'u da broda .i je da brode} = {da zo'u ge da broda gi da brode} = {suzdza lo ka poi'i ge ke'a broda gi ke'a brode}
**05:03 < Ilmen> {da poi'i ge ke'a broda gi ke'a brode}   
**05:03 < Ilmen> {da poi'i ge ke'a broda gi ke'a brode}   
**1. and 2. are to be solved by using prototypes. {{jvs|xunre}} is an idealistic prototype not present in real world. Or we can remember Einstein's words that all electrons have the same mass because they are simply the same electron (projected many times into the physical Universe).
13:41 < selpahi> {ze'i so'i roi plipe} vs {so'i roi ze'i plipe}, {ze'i lo mentu mi so'i roi [ri] cu plipe} vs {so'i roi ku mi plipe ze'i lo snidu}
*22:34 < gleki> 1. there are 100 women, 10% of them are pretty. He likes 10 those women
 
22:35 < gleki> 2. There are 100 women, 10 % of them are pretty. He like 10% of pretty women, i.e. he likes one woman.
{puku da renvi} = there were survivors.
22:36 < durka42> remai nelci dauce'i lo ro ninmu poi melbi ku noi panomei
 
22:37 < durka42> camxes: -f nelci dauce'i lo ro ninmu poi melbi ku noi panomei
{da puku renvi} = Some survived.
22:37 < camxes> (nelci [dau ce'i] [lo {ro ninmu <poi melbi>} ku] [noi {pa no} mei])
 
22:38 < durka42> panono da ninmu .i pano ri melbi .i nelci ri
{mi co'u ponse da} ----> tolcfa fa lo nu li su'o kaidza lo ka mi ponse ke'a - i lost everything.
22:38 < durka42> ja'o nelci ro lo pano lo panono ninmu ku poi melbi
{da co'u se ponse mi} - there is something i lost.
22:39 < durka42> camxes: -f ja'o nelci ro lo pano lo panono ninmu ku poi melbi
 
22:39 < camxes> (ja'o [nelci {ro <lo (¹pa no¹) (¹lo [{pa <no no>} ninmu] ku¹) (¹poi melbi¹)>}])
I wasn't sure whether you were saying you come here when you are writing, or you don't come here often because you are usually writing
22:39 < durka42> replace {lo pano} with {lo dauce'i} if you want
 
22:40 < maik_> gleki, you meant to say "He likes _the_ 12% of women who are pretty" earlier
{mi ru'inai jundi mu'i lo nu finti} implies the former
*12:44 < gleki> how to express "if(broda){brode}else{brodi}" without repeating broda twice?
----
12:45 < b_jonas> tsanire: it is the start of {.i ga nai go'i gi brodu}
"i come here rarely in order to write" - "in order to write i come here rarely"
12:45 < b_jonas> tsanire: so the whole thing would be {.i brodi .i ja broda .i ga nai go'i gi brodu}
----
12:46 < b_jonas> and I'm asking whether that means the same as {.i brodi .i ja broda .i ga nai broda gi brodu}
{mi na klama mu'i tu'a lo carvi} (“It is not true that [I go motivated by the rain]”), when they actually mean {mi na'e klama mu'i tu'a lo carvi} (“[I don't go], which is motivated by the rain”).
12:47 < b_jonas> and if it doesn't mean that, then I wonder what you could say instead of "go'i" to refer back to just "broda", where "broda" here is placeholder for a whole bridi which could be long
 
12:49 < latro`a> if there's some dispute about boundaries between bridi, then you could circumlocute by reifying the selbri and assigning it
< durka42> gleki: the classic example of going to the bank in the rain, for one
12:49 < latro`a> {.i brodi .i ja me'au lo du'u broda goi ko'a .i ga nai me'au ko'a gi brodu}
10:30 < gleki> durka42: pardon?
12:49 < b_jonas> latro`a: yes, you could certainly assign it with cei (if I understand it right what cei means)
10:31 < lai_az> le nu mutce nandu mi cu rinka zo'e bi'unai
12:50 < b_jonas> I don't think you have to reify it for that
10:31 < lai_az> or zo'e bi'u if its new
12:50 < latro`a> if go'i has a problem than cei does too-
10:31 < lai_az> or ti etc
12:50 < b_jonas> just assign it to brodo with cei
10:31 < durka42> gleki: mi na klama lo banxa ki'u lo nu carvi
12:50 < latro`a> they're essentially the same idea
10:32 < gleki> sounds like "i dont drink alcohol for religious reasons"
12:50 < b_jonas> what... oh cr
10:32 < durka42> same, yes
12:50 < b_jonas> you're right
10:33 < gleki> but how to translate it? for me the straightforward translations is a mental puzzle
12:50 < b_jonas> it's because cei is too greedy
10:33 < durka42> na-ki'u: religion isn't the reason I drink alcohol; ki'u-na: due to religion I don't drink
12:50 < b_jonas> it gobbles up more than you put it next to
10:46 < gleki> and with the rain?
* < Ilmen> it's similar to selpa'i's proposal for a cmavo allowing "proprietor fronting" (split of a NU into an object and a property)
10:46 < durka42> na-ki'u: rain doesn't draw me to the bank, or something, u'i
09:15 < mukti> So the "part of" bit of your gloss of {jai cumki} has me interested.
 
09:15 < danr> mukti: It should be an easily accesible sumti
Following Lojban usage, I observe an extremely high number of scope mistakes. It appears that people are generally unaware. Their sentences are usually still understood "thanks" to the human capability of letting context fix scope dependencies. That's why these mistakes tend to go unnoticed. Nobody corrects these mistakes, and they keep on being made. It's a problem, but it's not too late to get a new awareness of scope.One of the most common scope mistake is this:
09:16 < mukti> Somehow x1 participates in xFAI ?
 
09:16 < Ilmen> "he's easy to trick" ---> ko'a jai frili loka se tcica
(1) mi na terve'u lo stagi ki'u lo nu mi na bevri lo jdini
09:16 < gleki> quickly spltting an abstraction into an object+ its property is one of the highly needed features in lojban.
"I did not buy vegetables because I didn't have money with me."
09:16 < Ilmen> in kind of nu-split is common in English
 
*«lu no da zo'u: mi smàdi lo du'u da dùnda lo jdìni li'u» frìca «lu mi smàdi lo du'u no da zo'u: da dùnda lo jdìni li'
If you just read the English, your mind is likely to autocorrect the scope, causing you to miss the problem in the Lojban sentence. Though this might seem like an obvious mistake, I just saw someone make it again today.
*{fi'o gunma je fa lo tadni cu sruri}?
 
If ca'a modifies a selbri, why not inside a lo construct?
Sentence (1) actually means:
08:05 < latro`a> darsi: that would modify the wrong selbri
"That I bought vegetables because I didn't have money with me is false."
08:05 < darsi> oh, right.
 
08:05 < latro`a> we want {ko'a ca'a mruli .i ko'e ka'e se mruli} and then to somehow join these two
That's how the Lojban should be understood, but it often isn't. A practical fix is to use {.i ki'u bo}:
* no, I don't want erasure erasure is textual. I'd like semantic one, not syntactic erasure                         
 
23:51  erasure with {sa} is ok if I misspoke, but in this case I didn't misspoke, I was just wrong about some fact                   
(2) mi na terve'u lo stagi .i ki'u bo mi na bevri lo jdini
23:51 < gleki> {mi bebna i go'i je'unai i do go'i} = I am a fool. Wait, it's not true. YOU are a fool.                                             
"I didn't buy vegetables. The reason for that is that I didn't have money with me."
23:52  gleki: yes, that negates the previous statement, but then how do I tell the next statement is something of similar scope that corrects the wrong statement, not just something unrelated?                                                           
 
23:53 < gleki> I need an example because .uinai I cant understand you yet.           
The other method requires forethought, which might seem annoying, but you'll soon come to realize (if you haven't already) that forethought is needed in all of Lojban the moment you begin to scope-juggle.
23:55  ok, how about something like "The house is big. No, that's wrong. Actually, the house is small."                               
 
23:56  now {.i go'i .uinai} could express the "No, that's wrong." part, but how do I say "Actually,"?                                 
(3) mi ki'u lo nu na bevri lo jdini cu na terve'u lo stagi
23:58 < gleki> lo dinju cu barda i oi go'i je'unai i lo go'i cu cmalu je'u            
"I, because [I] didn't have money with me, did not buy vegetables."
Day changed to 27 Mar 2014                                                           
00:00 < gleki> It might look tricky because {lo go'i} refers to the first sumti in the second sentence. And {go'i} in the second sentence refers to the phrase in the first sentence.
*You can't start a run of sumti with ke, for reasons of Lojban grammatical pedantry we won't go into here. WTH?
*a prosumti for something actually unknown would be cool
**13:31 < acolotl> E.g. {mi broda Y} -> {mi broda zo'e noi ke'a ka'e se no'a}
**13:31 < acolotl> It can be used for a restricted "everything"
**13:32 < acolotl> {mi djuno ro da} could be understood as too extreme
**13:32 < acolotl> but {mi djuno ro Y} is "I know everything that is knowable"
**13:33 < acolotl> And {ro Y prami do} means "Everyone that can love you does"
**13:33 < acolotl> Y is just everything that could fill this place
**13:33 < xalbo> I think it's just more useful to assume that {da} is contextually bound, because otherwise it's completely useless.
*All languages need ways to connect words, phrases and sentences. In English there are a host of words for this purpose: ''and, or, because, additionally, however, on the other hand'' ... the list seems endless, as foreign students of English know all too well.
**Lojban also has a wide variety of words like this, known as ''conjunctions'', but it is more systematic about it.
**There are two types of conjunctions: logical and non-logical. Logical conjunctions say something about whether and in what circumstances the two things connected are true; an example is '''je'''. Non-logical conjunctions do not deal with separate truth values, but group things together to form different kinds of units.
**Lojban distinguishes between the logical component of conjunctions, and their attitudinal content. For example, most languages have different words for ''and'' and ''but''. Logically, they both mean the same thing. In terms of attitude, however, they are different: ''but'' contains a connotation of contrast or unexpectedness, which ''and'' does not. So Lojban translates ''but'' in two parts: '''je ku'i''' = ''and however''. This follows the Lojban principle of keeping content and attitude separate (e.g. '''.ui la .jasmin. cu klama ti''' has a content element — the information that Jasmine is coming here, and an attitude element — happiness).
**Lojban also handles some of the functions of English conjunctions in other ways — as we saw, ''because'' and ''so'' are translated with prepositions, not conjunctions.
*I understand now that that is how it works, but if "ro" and "pa" combine with other members of PA differently, then how can they be considered members of the same selma'o?
**They behave the same way grammatically, but their interpretation, which is a semantic feature not governed directly by the grammar, depends on the exact words used. The CLL describes these interpretations:
**Another possibility is that of combining definite and indefinite numbers into a single number. This usage implies that the two kinds of numbers have the same value in the given context:


8.18)  mi viska le rore gerku
Case in point?
      I saw the all-of/two dogs.
      I saw both dogs.


8.19)  mi speni so'ici prenu
There are also cases of scope mistakes that I observe even among otherwise experienced Lojbanists. This is usually the case when multiple tenses are being used. Often, people just tack on an additional tense at the end of their already-tensed phrase, and the result is usually a sentence with incorrect scope, which the speaker doesn't notice.
      I am-married-to many/three persons.
      I am married to three persons (which is “many” in the circumstances).


(Chapter 18, section 8, at the bottom.)
There is a difference between {ca pu} and {pu ca}. There is a difference between {pu co'a} and {co'a pu}. You normally cannot reverse the order of phrase operators and expect to get the same result each time. While initially a difficulty, it's also one of the strong points of Lojban. It's the very thing Lojban is supposed to be good at. To neglect this aspect is to neglect Lojban.


Some lojbanists extend these interpretations in many different directions, but those usages are (sadly) nonstandard. The general rule that we can discern from this CLL excerpt is that when an exact number and an inexact number are sequenced, then the result is as if these numbers were logically connected with {.e}.
Learning to use this feature correctly requires one to rethink. Lojban is not English. It's not even close. It just happens to sometimes be able to mirror the word order. It quickly stops doing that once the statements get more complex, at which time it drifts more towards SOV languages. It's not unnatural for a language to require constant forethoughts. Japanese and Turkish do it, just to name two. (Also note that when I say "forethought", I'm not talking about forethought connectives, although they can help. I'm talking about the general concept of thinking ahead).
*nau bu = lo cabna je zvati be lo nu cusku dei
 
**nau bu goi lo jai nau bu'a nau zu'i
Simply put, you have to learn (practically by rote, or by making the same mistake many times and correcting it until it sticks) what operators should precede what operators in what situation. Having some mental categories of different situations is very helpful and, I assume, inevitable. Well, to be sure, there are always other methods, but this one doesn't seem so painful. To give you an example, you need to simply know that a prenexed quantifier is required in this general instruction manual sentence:
*fu'e for nai
 
*lo du'u ti xunre cu bridi lo ka xunre kei ti {lo du'u} is "the fact that", except "the fact that 2+2=5" is silly and wrong, but {lo du'u li mu sumji li re li re} is fine, albeit false. You can think of it as "respectively"
(4) "Thirdly, everyone holds on to the rope."
*PUrci, CAbna, BAlvi
 
*we have bridi->adjective (NOI), bridi->selbri (NU), but no bridi->adverb
Suppose this is a sentence that describes a hypothetical setup that the instruction manual is trying to explain to you. We can argue that the rope is not definite. It can be any rope (so {lo} fits more than {le}), but everyone is supposed to hold on to the same rope, that's the important part. The naive translation fails:
*Explain '''lo nu broda pu brode''' - and use {lo nu broda cu pu brode}. so {lo} and {sei} require the selbri to be the last word. {nu} allows anything: prepositions, nouns, at the end
 
**'''lo nu tcidu ca nandu''' - ''The current reading complication''
(4a) ci mai .ei ro lo prenu cu jgari lo skori
**'''lo nu tcidu cu ca nandu''' - ''The reading is now difficult''
 
**'''lo nu tcidu ca cu nandu''' - ''The current reading is difficult.''
This doesn't tell us that everyone is supposed to hold on to the same rope. And we also cannot use {le} (and even if we could, it's not really the point of this example). (4a) is vague about the rope business. So what now? As I said, this type of situation needs a prenexed quantifier:
* xalbo: In your section "the simplest sentences", it might be beneficial to make one of your examples an "adjective" (colors are good), to show that they're verbs too.
** gleki: wait, adjectives are NOI
*** [20:26] xalbo: I mean using {blanu} in addition to {mlatu} and {pinxe}
**** [20:26] gleki: lo blabi mlatu cu pinxe lo ladru?
***** [20:27] xalbo: I'm saying include {blanu}, which is simply translated as "blue", but is really also a verb "to be blue".


=== banzu ===
(4b) ci mai .ei pa da poi skori zo'u: ro lo prenu da jgari
Standard:
 
*banzu = x1 (object) suffices/is enough/sufficient for purpose x2 under conditions x3.
This tells us explicitly that each person is supposed to hold on to the same rope.
*sarcu = x1 (abstract) is necessary/required for continuing state/process x2 under conditions x3.
*dukse = x1 is an excess of/too much of x2 by standard x3.
Abandoned proposal:
*banzu = x1 (nu) suffices/is enough/sufficient for purpose x2
*sarcu = x1 (nu) is necessary/required for continuing state/process x2
*dukse = x1 (nu) is an excess of/too much for continuing state/process x2
*Math:
and sarcu? how would you say x1 is necessary and sufficient?
10:43 < latro`a_> sarcu gi'e se jalge
10:46 < latro`a_> or in a more material implication sense, jalge gi'e se jalge
10:46 < latro`a_> which is basically the nu version of {jo}
10:46 < latro`a_> but then, I haven't seen "necessary and sufficient" in an informal context in a long, long time
10:47 < latro`a_> whereas in a formal context {jo} is sufficient
**New:
***sufsi = x1 is enough in x2 (ka/ni) for purpose x3 (nu) to take place
***dukse = x1 is too much in x2 (ka/ni) for purpose x3 (nu) to take place
the ka form is the same as the ni form after substitution, but with dukse2/banzu2 being zo'e
10:54 < latro`a_> for example, {mi dukse lo ka ma kau zdani ce'u} = {lo zdani be mi cu dukse zo'e}
10:54 < latro`a_> but with modified emphasis
***barda and friends do the same


=== Super-Advanced: Negation ===
Yes, you can also just reverse the order of the sumti as in (5), but it would be a different kind of word order (which proves my point, but messes up the example. If you want to mirror the English, (4b) is the way to go).
Lojban presents several words for expressing different shades of negation.
==== na ====
#'''na''' put immediately before the main verb negates the whole clause:
#:{{mupli|'''lo speni be mi cu na ninmu'''<br/>It is not true that my spouse is a woman.}}
#::'''speni''' = ''x1 is married to x2''
#:It states nothing about what my wife is, or if I even have a wife. It only states that I do not have a wife who is also a woman.
#:Similarly, if someone asks you '''xu do pu viska la .ian.''' (''Did you see Yan?'') you can reply '''na go'i''' (''No.'') even you have never heard of any Yan in your life. And you won't be really lying.
==== na ku ====
#'''na ku''' in the beginning of a clause negates the whole clause:
#:{{mupli|'''na ku lo speni be mi cu ninmu''' and '''lo speni be mi cu na ninmu''' mean the same.<br/>It is not true that my spouse is a woman.}}
#'''na ku''' in other positions negates everything to the right of itself within the current clause.
#:If I say '''mi na sutra tavla bau lo glibau se ja'e lo nu mi dotco''', I end up negating too much, and it is not clear that I wanted to only negate that I speak fast. The sentence could suggest that I in fact speak fast because of some other reason, for instance that I speak fast in French because I'm German. To express the sentence more precisely, I need to only negate that I speak fast, and not the other things. To only negate part of a clause, '''na ku''' can be moved around the clause and placed anywhere a noun can go. It then negates any noun, main verb and preposition placed after it.
#'''na ku''' at the rightmost end of the clause is the same as '''na''' before the the main verb of this clause.
#:Often we need our negation to be a little less powerful. In particular, it is useful to be able to say, not that the whole clause is false, but only the main verb. This means that there <u>is</u> some relationship between the nouns — but this main verb isn't it.


Moving '''na ku''' from the left end of the sentence and rightwards effects any quantifiers in a certain way, as can be seen by this example:
(5) ci mai .ei pa skori cu se jgari ro lo prenu
{{mu|na ku ro lo remna cu verba|It's not true that: All humans are children.}}
{{mu|su'o lo remna na ku cu verba|For at least one human it's not true that: it's a child.}}
See that the '''na ku''' is placed before '''cu''', since a noun can go only before, not after the '''cu'''. Had I only used '''na''', it would have to go after '''cu''' — but that would have negated the entire clause, meaning "It's not true that: At least one human is a child".


When the '''na ku''' is moved rightwards, any quantifier is inverted — that is: '''ro''' is turned into '''su'o'''. This is, of course, only if the meaning of the clause has to be preserved. This means that when the '''na ku''' is placed at the end of the clause, only the main verb is negated but all the nouns and prepositions are preserved, as can be seen by these three identical clauses:
This one is shorter, not requiring an explicit prenex, and no explicit {da} binding. It does require one to change the order of the arguments, though, which might not always be desirable. Sometimes, placing a sumti in the x1 achieves a certain effect, e.g. topicalization and switching arguments can potentially ruin that.


{{mu|na ku ro lo verba cu tadni bu'u da poi ckule|It's not true that all children are students in a school.}}
It's also handy to know that generally location and time tags come before everything else, but you need to know in what types of situations they don't. For instance, try to figure out when you want {ca ... pu ...} and when {pu ... ca ...}.
{{gl|ckule|x1 is a school at area x2}}


{{mu|su'o lo verba cu tadni na ku bu'u da poi ckule|Some children study in not a single school.}}
Raise your awareness. But beware, you'll suddenly notice lots of mistakes! Maybe you don't want to eat of that tree.
{{mu|su'o lo verba cu tadni bu'u ro lo ckule na ku|Some children are for all schools don't study in them.}}


It is possible to use clause negation in all clauses, even the implicit clause of descriptive nouns. '''lo na prenu''' can refer to anything non-human, whether it be a sphinx, a baseball or the property of appropriateness.
Anyway, in this fashion every possible scenario can be put into a category.


{{notci|Note: Like '''cu''', '''na''' can indicate that a main verb is coming up, so we can just say '''na''' instead of '''cu na'''. So you can say
I'm not going to lay out all those possibilities here. I think it's good to think about this topic for yourself and to compare the different meanings the different scopes produce.
:'''lo ninmu na nelci la .ian.''' {{=}} ''A woman likes Yan''
without adding '''cu''' before '''na nelci'''.}}


==== Interjections '''nai''', '''ru'e''', '''sai''', '''cu'i''' and '''na'i''' ====
Seeing so many scope mistakes every day is slightly painful for me, because I currently have a very high awareness of the subject matter. It's easy to overlook, and I wasn't always aware of it either. Heck, only recently did I realize that my staple {tai ... ja'e ...} has inccorect scope. I'm sure there are many more constructions out there that need to be reconsidered. I'm not afraid to re-adjust my Lojban, and I recommend everyone to do the same.
#'''nai''' being an interjection attaches to the right of a construct and negates it.
#:{{mupli|'''tu bajra mlatu nai'''<br/>''That is a running non-cat.''}}
#:{{mupli|'''tu bajra nai mlatu'''<br/>''That is a non-running cat.''}}
#:As you can see '''nai''' is applied only to the last verb of a compound verb.
#When attempting to answer: “Is the king of the USA fat?”, all of these negations fail. While it's technically correct to negate it with na, since it makes no assumptions of that is true, it's mildly misleading since it could lead the listener to believe there is a king of the USA. For these scenarios, there is a metalinguistic negator, '''na'i'''.  
#:'''na'i''' — metalinguistic negator. Something is wrong with assigning a truth value to the clause.
#:Because '''na'i''' has the grammar of interjections.
#:{{mupli|'''lo na'i pu te zukte be lo skami cu palci'''<br>''The sought goal (mistake!) of the computer was evil''.
#:'''palci''' = ''x1 is evil''
#:This sentence protests that computers can seek a goal volitionally.}}


==== Scalar negation ====
Lojban tends to be left-grouping, tanru and connectives for example are left-grouping. However, scope is ... "right-grouping". I'm not sure if others conceptualize it this way, but one way to parse multiple scopes is to consider each scope operator to create a sort of bubble to its right, and those bubbles are right-grouping. Abstractly:
'''na'e''' attaches to the left of a construct and negates it implying the existence of a scale:
Abstractly:
{{mu|tu na'e bajra mlatu|That is a other-than-running cat.}}
<pre>⎛              ⎛                      ⎛          ⎞⎞⎞
{{mu|tu bajra na'e mlatu|That is a running other-than-cat.}}
.i op1⎜  A  op2 ⎜  broda  B  op3⎜  C  D  ⎟⎜⎜
:(making us wonder what a non-cat might be)
⎝                  ⎝                      ⎝          ⎠⎠⎠
:{{mupli|''My spouse is not a woman'' (meaning that he is a male).<br>'''lo speni be mi cu na'e ninmu''' or '''lo speni be mi cu to'e ninmu'''.
:Using scalar negation here implies that he exists, which '''na''' did not.}}


Another unexpected effect compared to English ''not'':
</pre>
:{{mupli|'''mi na nelci ro lo gerku''' means ''It is not true that I like each dog''.}} However,
op = phrase operator,  capital letters represent constants (i.e. unquantified sumti), the brackets should speak for themselves.
:{{mupli|'''mi na'e nelci ro lo gerku''' means ''I don't like any dogs.''}}
While the mechanism of '''na ku''' resembles negation in English, it can be difficult to keep track of exactly what is negated and how that affects the clause. For that reason, the construct '''na ku''' is rarely seen anywhere other than the beginning of a clause. In most cases where more specific negation is needed people resort to a scalar negation using '''na'e'''. With '''na'e''' only a ceratin construcut is affected whereas with '''na ku''' everything to the right of it is negated. This is the reason '''na ku''' is used rather seldom and mostly at the beginning of the sentence. In
:{{mu|mi <u>na'e</u> nelci ro gerku|I other-than-like all dogs.}}
there is something that can be said about me and all dogs, but it's not that I like them.


I might hate them:
Obviously, I think scope should be given a lot more importance. It is the thing that determines the structure and word order of sentences. By acknowledging it, we can get closer to Lojban's true word order.
{{mu|mi <u>to'e</u> nelci ro gerku|I anti-like ({{=}} dislike) all dogs.}}


'''to'e''' turns a main verb into its opposite: '''to'e nelci''' is pretty much the same thing as '''xebni''' — ''to hate''.  
=== Issues: unsorted ones ===
 
**1. and 2. are to be solved by using prototypes. {{jvs|xunre}} is an idealistic prototype not present in real world. Or we can remember Einstein's words that all electrons have the same mass because they are simply the same electron (projected many times into the physical Universe).
Or I just might be neutral to them. I might write poems about them, or prescribe medicine for them, imitate them in polite company, or just might be indifferent to them. Then we'll use '''no'e''':
*22:34 < gleki> 1. there are 100 women, 10% of them are pretty. He likes 10 those women
{{mu|mi <u>no'e</u> nelci ro gerku|I am neutral-as-to-liking all dogs.}}
22:35 < gleki> 2. There are 100 women, 10 % of them are pretty. He like 10% of pretty women, i.e. he likes one woman.
But as for liking them, I don't.
22:36 < durka42> remai nelci dauce'i lo ro ninmu poi melbi ku noi panomei
 
22:37 < durka42> camxes: -f nelci dauce'i lo ro ninmu poi melbi ku noi panomei
Thus '''na'e''' is not a negator in the same sense as '''na'''. It doesn't state that a clause is false, but makes a positive statement that a clause is true – the same clause, but with a different main verb. This distinction is mostly academic, though. If, for example, I state '''mi na'e se nelci''' — ''I am non-liked'', I actually state that some main verb applies to me, which is also on a relevant scale with the main verb '''nelci'''. Most of the time, we assume a scale where the positions are mutually exclusive like "like — dislike — hate", so '''mi na'e se nelci''' implies '''mi na se nelci'''. Therefore, the words '''no'e''' and '''to'e''' should only be used when the main verb is placed on some obvious scale: '''lo speni be mi cu to'e melbi''' ''My spouse is ugly'' makes sense, since we immediately know what the opposite of beautiful is, while '''mi klama lo to'e zdani be mi''' – ''I go to my opposite thing of home'', while grammatical, leaves the listener guessing what the speaker's ''opposite of home'' is and should be avoided.
22:37 < camxes> (nelci [dau ce'i] [lo {ro ninmu <poi melbi>} ku] [noi {pa no} mei])
:{{mupli|''My spouse is not really a woman.''<br>'''lo speni be mi cu no'e ninmu'''}}
22:38 < durka42> panono da ninmu .i pano ri melbi .i nelci ri
The scale here is presumed to be from woman to man.
22:38 < durka42> ja'o nelci ro lo pano lo panono ninmu ku poi melbi
:{{mupli|''I don't speak fast in English because I'm German.''<br>'''mi na'e sutra tavla bau lo glibau se ja'e lo nu mi dotco'''}}
22:39 < durka42> camxes: -f ja'o nelci ro lo pano lo panono ninmu ku poi melbi
 
22:39 < camxes> (ja'o [nelci {ro <lo (¹pa no¹) (¹lo [{pa <no no>} ninmu] ku¹) (¹poi melbi¹)>}])
==== Chat ====
22:39 < durka42> replace {lo pano} with {lo dauce'i} if you want
<pre>
22:40 < maik_> gleki, you meant to say "He likes _the_ 12% of women who are pretty" earlier
05:05 < plise> i ma drani gerna fi zo na
*12:44 < gleki> how to express "if(broda){brode}else{brodi}" without repeating broda twice?
05:05 < plise> i ma drani gerna fi zo na fe lo cabna jbobau
12:45 < b_jonas> tsanire: it is the start of {.i ga nai go'i gi brodu}
05:05 < plise> sisi balvi jbobau
12:45 < b_jonas> tsanire: so the whole thing would be {.i brodi .i ja broda .i ga nai go'i gi brodu}
05:05 < markuja> ta'i ma mi spuda
12:46 < b_jonas> and I'm asking whether that means the same as {.i brodi .i ja broda .i ga nai broda gi brodu}
05:05 < markuja> .i xu jarco lo me PEG javni
12:47 < b_jonas> and if it doesn't mean that, then I wonder what you could say instead of "go'i" to refer back to just "broda", where "broda" here is placeholder for a whole bridi which could be long
05:06 < plise> go'i
12:49 < latro`a> if there's some dispute about boundaries between bridi, then you could circumlocute by reifying the selbri and assigning it
05:06 < plise> i ku'i au zgana lo balvi jbobau
12:49 < latro`a> {.i brodi .i ja me'au lo du'u broda goi ko'a .i ga nai me'au ko'a gi brodu}
05:06 < markuja> .oi
12:49 < b_jonas> latro`a: yes, you could certainly assign it with cei (if I understand it right what cei means)
05:06 < plise> y i ku'i e'a do ciksi bau lo glico
12:50 < b_jonas> I don't think you have to reify it for that
05:06 < markuja> .i xu do na kakne lo ka catlu
12:50 < latro`a> if go'i has a problem than cei does too-
05:06 < plise> i mi nitcu tu'a lo mi ctuca cukta
12:50 < b_jonas> just assign it to brodo with cei
05:06 < nargavela> BTW I should ask on the mriste how to express the PEG '/' operator; it is a must for describing Lojban syntax in Lojban
12:50 < latro`a> they're essentially the same idea
05:06 < markuja> .e'a nai glico
12:50 < b_jonas> what... oh cr
05:06 < plise> i mi kanpe lo nu da ba fanva CC lo porto i seki'u bo mi xanka
12:50 < b_jonas> you're right
05:07 < markuja> ma pe zo na nabmi
12:50 < b_jonas> it's because cei is too greedy
05:07 < markuja> .i pe'i zo na no'e cenba
12:50 < b_jonas> it gobbles up more than you put it next to
05:07 < plise> markuja: xu dunlo zo na pe cylyly
* < Ilmen> it's similar to selpa'i's proposal for a cmavo allowing "proprietor fronting" (split of a NU into an object and a property)
05:07 < plise> dunli
09:15 < mukti> So the "part of" bit of your gloss of {jai cumki} has me interested.
05:07 < markuja> gerna dunli
09:15 < danr> mukti: It should be an easily accesible sumti
05:07 < gerna> (0[dunli VAU])0
09:16 < mukti> Somehow x1 participates in xFAI ?
05:07 < markuja> .i gerna dunli
09:16 < Ilmen> "he's easy to trick" ---> ko'a jai frili loka se tcica
05:07 < nargavela> ta'o .ei mi retsku fi selmristepre fe su'u makau xe fanva zo'oi / pe PEG
09:16 < gleki> quickly spltting an abstraction into an object+ its property is one of the highly needed features in lojban.
05:08 < plise> markuja: xu gerna mintu zo na pe CLL
09:16 < Ilmen> in kind of nu-split is common in English
05:08 < markuja> go'i
*«lu no da zo'u: mi smàdi lo du'u da dùnda lo jdìni li'u» frìca «lu mi smàdi lo du'u no da zo'u: da dùnda lo jdìni li'u»
05:08 < markuja> .i ku'i mi curmi lo nu zo ku na se cusku
*{fi'o gunma je fa lo tadni cu sruri}?
05:08 < plise> mi tugni la ionac lo nu tai da'i la'edi'u zo fa ba plujygau lo gerna
If ca'a modifies a selbri, why not inside a lo construct?
05:09 < plise> markuja: xu lu i mi na broda  li'u smuni  mintu lu i naku mi broda li'u
08:05 < latro`a> darsi: that would modify the wrong selbri
05:09 < markuja> no'e go'i
08:05 < darsi> oh, right.
05:10 < plise> i mintu ma
08:05 < latro`a> we want {ko'a ca'a mruli .i ko'e ka'e se mruli} and then to somehow join these two
05:10 < nargavela> lo se stidi pe ma'oi fa na dirba mi .i ku'i la markuja pu cpedu fi tu'a lo pilno mupli
* no, I don't want erasure erasure is textual. I'd like semantic one, not syntactic erasure                         
05:10 < markuja> lu mi broda na ku li'u
23:51  erasure with {sa} is ok if I misspoke, but in this case I didn't misspoke, I was just wrong about some fact                   
05:10 < plise> uanai
23:51 < gleki> {mi bebna i go'i je'unai i do go'i} = I am a fool. Wait, it's not true. YOU are a fool.                                             
05:10 < nargavela> y .i lo nunpli mupli
23:52  gleki: yes, that negates the previous statement, but then how do I tell the next statement is something of similar scope that corrects the wrong statement, not just something unrelated?                                                           
05:10 < markuja> doi plise
23:53 < gleki> I need an example because .uinai I cant understand you yet.           
05:10 < markuja> ju'i plise
23:55  ok, how about something like "The house is big. No, that's wrong. Actually, the house is small."                               
05:10 < plise> markuja: mi mutce tatpi doi sy i e'o do ciksi do'e mu'a lo do blogo
23:56  now {.i go'i .uinai} could express the "No, that's wrong." part, but how do I say "Actually,"?                                 
05:10 < markuja> .i ko karbi lu na broda gi'e brode li'u
23:58 < gleki> lo dinju cu barda i oi go'i je'unai i lo go'i cu cmalu je'u            
05:11 < markuja> .i lu na ku broda gi'e brode li'u frica sai
Day changed to 27 Mar 2014                                                           
05:11 < markuja> ie pei
00:00 < gleki> It might look tricky because {lo go'i} refers to the first sumti in the second sentence. And {go'i} in the second sentence refers to the phrase in the first sentence.
05:11 < plise> i mi na birti lodu'u lu i broda naku li'u cu se smuni makau
*a prosumti for something actually unknown would be cool
05:11 < plise> y
**13:31 < acolotl> E.g. {mi broda Y} -> {mi broda zo'e noi ke'a ka'e se no'a}
05:11 < plise> ie
**13:31 < acolotl> It can be used for a restricted "everything"
05:11 < markuja> smuni fa lo du'u lo du'u broda na jetnu
**13:32 < acolotl> {mi djuno ro da} could be understood as too extreme
05:11 < nargavela> lu na broda gi'e brode li'u zo'u selbritcita
**13:32 < acolotl> but {mi djuno ro Y} is "I know everything that is knowable"
05:11 < plise> i na broda = i broda naku ?
**13:33 < acolotl> And {ro Y prami do} means "Everyone that can love you does"
05:12 < markuja> panra zo pu
**13:33 < acolotl> Y is just everything that could fill this place
05:12 < markuja> .i pu broda gi'e brode
**13:33 < xalbo> I think it's just more useful to assume that {da} is contextually bound, because otherwise it's completely useless.
05:12 < nargavela> na broda = broda be naku
*There are two types of conjunctions: logical and non-logical. Logical conjunctions say something about whether and in what circumstances the two things connected are true; an example is '''je'''. Non-logical conjunctions do not deal with separate truth values, but group things together to form different kinds of units.
05:12 < markuja> .i pu broda gi'e ba brode
**Lojban distinguishes between the logical component of conjunctions, and their attitudinal content. For example, most languages have different words for ''and'' and ''but''. Logically, they both mean the same thing. In terms of attitude, however, they are different: ''but'' contains a connotation of contrast or unexpectedness, which ''and'' does not. So Lojban translates ''but'' in two parts: '''je ku'i''' = ''and however''. This follows the Lojban principle of keeping content and attitude separate (e.g. '''.ui la .jasmin. cu klama ti''' has a content element — the information that Jasmine is coming here, and an attitude element — happiness).
05:12 < markuja> .i na broda gi'e ja'a brode
**Lojban also handles some of the functions of English conjunctions in other ways — as we saw, ''because'' and ''so'' are translated with prepositions, not conjunctions.
05:12 < plise> "naku" srana lo jufra, "na" srana lo bridi?
*I understand now that that is how it works, but if "ro" and "pa" combine with other members of PA differently, then how can they be considered members of the same selma'o?
05:12 < markuja> .i .o'i doi nargavela
**They behave the same way grammatically, but their interpretation, which is a semantic feature not governed directly by the grammar, depends on the exact words used. The CLL describes these interpretations:
05:13 < markuja> .i lu na broda su'o da li'u na dunli lu broda be naku su'o da li'u
**Another possibility is that of combining definite and indefinite numbers into a single number. This usage implies that the two kinds of numbers have the same value in the given context:
05:13 < nargavela> ie u'u
05:13 < plise> "naku" srana lo jufra, "na" srana lo bridi?
05:13 < nargavela> ki'e kajde
05:14 < markuja> zo na co'e lo pritu ke galfi valsi noi zo pu .e zo su'o mupli
05:14 < nargavela> lu naku li'u sumsmi
05:14 < markuja> .i zo na ku simsa
05:14 < markuja> sa
05:14 < markuja> .i lu naku li'u simsa
05:14 < markuja> .i ie
05:16 < plise> ei mi troci lo ka ciska ciksi
05:18 < nargavela> lu na broda li'u lu naku broda li'u frica va'o lo nu bridi       
 
                  konketiva                                                         
05:19 < markuja> mekso: -f broda su'o da na .i broda na su'o da                     
05:19 < mekso> ([broda {su'o da} na] [i {broda <na (¹su'o da¹)>}])
 
</pre>
 
==== Negation and neutral meaning ====
'''nai''' is an interjection that makes a part of sentence negative in meaning. '''nai''' means ''no'' or ''not''.
*interjection in Lojban modifies the construct to the left of it.
{{mu|mi nelci nai do|I don't like you.|I like-not you (literally).}}
{{mu|mi nai nelci do|Not I like you.|(may be someone else likes you)}}
:So when put after certain part of the phrase like pronoun or a verb it modifies that verb.
*if put in the beginning of a phrase interjection modifies the whole phrase:
{{mu|nai mi nelci do|It's not true that I like you.}}
:So in the beginning of the phrase '''nai''' negates the whole phrase.
*we can put an interjection after different parts of the same phrase shifting the meaning.
 
The interjection '''cu'i''' makes a part of sentence middle in its meaning.
{{mu|mi nelci cu'i do|As for whether I love or hate you, I'm indifferent to you. I neither like nor hate you.}}
 
==== Existential import ====
* http://www.lojban.org/tiki/On+the+meaning+of+%27ro+broda+cu+brode%27
* http://www.lojban.org/tiki/quantification+and+the+meaning+of+ro
 
==== Double negation ====
Sometimes it's necessary to use double negation. Let's show how this works using examples from English and Chinese.
 
The structure 非...不可 (fēi … bùkě) is one of the most commonly used in Mandarin Chinese. It means "must"/"absolutely must"/"need to." 非 means "not"/"no" and 不可 means "not possible". It's literally translated as "not not possible."
:'''mi na cfifa'i ra na fau da'''
:我非批评她不可。
:''wǒ fēi pīpíng tā bùkě''
:<code>I can't not to criticize her</code>.
:''I absolutely must criticize her.''


:'''mi'ai na tadni na fau da'''
8.18)  mi viska le rore gerku
:我们非学习不可。
      I saw the all-of/two dogs.
:''wǒmen fēi xuéxí bùkě''
      I saw both dogs.
:''We must study.''


:'''mi'ai tadni nai fau no da'''
8.19)  mi speni so'ici prenu
:<code>We can't not study.</code>
      I am-married-to many/three persons.
      I am married to three persons (which is “many” in the circumstances).


=== fau and if ===
(Chapter 18, section 8, at the bottom.)
There is some controversy about against using '''jo''' and '''ja nai''' for expressing ''if'' and ''only if''. First you shouldn't use '''ja nai''' (''if'') when '''jo''' (''only if'') is meant. And in natural languages ''if'' is strongly tied up with notions of causality, precondition, or deduction — none of which is particularly emphasized by '''ja nai''' and '''jo''' as they are strictly logical conjunctions. For example, '''ja nai''' will give a poor rendering of ''It's not true that, if I'm rich, I'm happy'' — which is decidedly not the same thing as ''It's not true that I'm either not rich or happy''! For that reason, you will see many Lojbanists avoiding '''jo''' and '''ja nai''' in ordinary speech, and instead using prepositions like '''fau''' - ''in the event of...'', '''va'o''' - ''under conditions...'', '''se ja'e''' - ''results from ... happening'', or '''ni'i''' - ''logically caused by...''.


Better examples of using them in ordinary speech:
Some lojbanists extend these interpretations in many different directions, but those usages are (sadly) nonstandard. The general rule that we can discern from this CLL excerpt is that when an exact number and an inexact number are sequenced, then the result is as if these numbers were logically connected with {.e}.
*Explain '''lo nu broda pu brode''' - and use {lo nu broda cu pu brode}. so {lo} and {sei} require the selbri to be the last word. {nu} allows anything: prepositions, nouns, at the end
**'''lo nu tcidu ca nandu''' - ''The current reading complication''
**'''lo nu tcidu cu ca nandu''' - ''The reading is now difficult''
**'''lo nu tcidu ca cu nandu''' - ''The current reading is difficult.''
* xalbo: In your section "the simplest sentences", it might be beneficial to make one of your examples an "adjective" (colors are good), to show that they're verbs too.
** gleki: wait, adjectives are NOI
*** [20:26] xalbo: I mean using {blanu} in addition to {mlatu} and {pinxe}
**** [20:26] gleki: lo blabi mlatu cu pinxe lo ladru?
***** [20:27] xalbo: I'm saying include {blanu}, which is simply translated as "blue", but is really also a verb "to be blue".


:'''ja gi nai je gi ro da poi badna zo'u da pelxu gi mi citka lo badna gi mi citka lo pelxu'''
=== banzu ===
:''If everything for every banana it is true that it is yellow, and I am eating a banana, then I am eating something yellow.''
Standard:
:'''ja gi nai je gi ro pipybanfi cu crino gi la .kermit. cu pipybanfi gi la .kermit. cu crino'''
*banzu = x1 (object) suffices/is enough/sufficient for purpose x2 under conditions x3.
:''If every frog is green, and Kermit is a frog, then Kermit is green.''
*sarcu = x1 (abstract) is necessary/required for continuing state/process x2 under conditions x3.
*dukse = x1 is an excess of/too much of x2 by standard x3.
Abandoned proposal:
*banzu = x1 (nu) suffices/is enough/sufficient for purpose x2
*sarcu = x1 (nu) is necessary/required for continuing state/process x2
*dukse = x1 (nu) is an excess of/too much for continuing state/process x2
*Math:
and sarcu? how would you say x1 is necessary and sufficient?
10:43 < latro`a_> sarcu gi'e se jalge
10:46 < latro`a_> or in a more material implication sense, jalge gi'e se jalge
10:46 < latro`a_> which is basically the nu version of {jo}
10:46 < latro`a_> but then, I haven't seen "necessary and sufficient" in an informal context in a long, long time
10:47 < latro`a_> whereas in a formal context {jo} is sufficient
**New:
***sufsi = x1 is enough in x2 (ka/ni) for purpose x3 (nu) to take place
***dukse = x1 is too much in x2 (ka/ni) for purpose x3 (nu) to take place
the ka form is the same as the ni form after substitution, but with dukse2/banzu2 being zo'e
10:54 < latro`a_> for example, {mi dukse lo ka ma kau zdani ce'u} = {lo zdani be mi cu dukse zo'e}
10:54 < latro`a_> but with modified emphasis
***barda and friends do the same


==== Non-logical conjunctions ====
=== Super-Advanced: Negation ===
:'''lo jisra joi lo djacu''' - ''the juice and the water, considered together''
Lojban presents several words for expressing different shades of negation.
This is the same as '''lo gunma be lo jisra je lo djacu'''.
==== na ====
 
#'''na''' put immediately before the main verb negates the whole clause:
:'''djacu''' = ''x1 is some water''
#:{{mupli|'''lo speni be mi cu na ninmu'''<br/>It is not true that my spouse is a woman.}}
:'''jisra''' = ''x1 is some juice''
#::'''speni''' = ''x1 is married to x2''
Not all English sentences containing ''and'' are like this, though. Firstly, sentences like ''I had a bath and washed my hair'' are structurally different and will be dealt with later on. Secondly, ''I visited Ranjeet and Jasmine'' is slightly different from ''I visited Ranjeet <u>and</u> I visited Jasmine.'' In this case, you probably want to say that you visited Ranjeet-and-Jasmine ''as a unit'' on one occasion — not that you visited Ranjeet and Jasmine on (potentially) different occasions (''It is true that I visited Ranjeet, and it is true that I visited Jasmine.'') In this case you don't want '''je''' (which is true but potentially misleading), but '''joi''', which means ‘in a mass with’. So what you have is
#:It states nothing about what my wife is, or if I even have a wife. It only states that I do not have a wife who is also a woman.
:'''mi pu vitke la .ranjit. joi la .jasmin.'''
#:Similarly, if someone asks you '''xu do pu viska la .ian.''' (''Did you see Yan?'') you can reply '''na go'i''' (''No.'') even you have never heard of any Yan in your life. And you won't be really lying.
:''I past visit Ranjeet in-a-mass-with Jasmine.''
==== na ku ====
:''I visited Ranjeet and Jasmine (together).''
#'''na ku''' in the beginning of a clause negates the whole clause:
#:{{mupli|'''na ku lo speni be mi cu ninmu''' and '''lo speni be mi cu na ninmu''' mean the same.<br/>It is not true that my spouse is a woman.}}
#'''na ku''' in other positions negates everything to the right of itself within the current clause.
#:If I say '''mi na sutra tavla bau lo glibau se ja'e lo nu mi dotco''', I end up negating too much, and it is not clear that I wanted to only negate that I speak fast. The sentence could suggest that I in fact speak fast because of some other reason, for instance that I speak fast in French because I'm German. To express the sentence more precisely, I need to only negate that I speak fast, and not the other things. To only negate part of a clause, '''na ku''' can be moved around the clause and placed anywhere a noun can go. It then negates any noun, main verb and preposition placed after it.
#'''na ku''' at the rightmost end of the clause is the same as '''na''' before the the main verb of this clause.
#:Often we need our negation to be a little less powerful. In particular, it is useful to be able to say, not that the whole clause is false, but only the main verb. This means that there <u>is</u> some relationship between the nouns — but this main verb isn't it.


This is just like the difference between '''lo ci gerku''' and '''loi ci gerku''' which we looked at in [[#4|Lesson 4]] — considering the three dogs as individuals, or as a mass. Incidentally, it is not just Lojban which makes this distinction; Turkish, for example, would use ''ile'' (‘with’) rather than ''ve'' (‘and’) for '''joi''' here.
Moving '''na ku''' from the left end of the sentence and rightwards effects any quantifiers in a certain way, as can be seen by this example:
{{mu|na ku ro lo remna cu verba|It's not true that: All humans are children.}}
{{mu|su'o lo remna na ku cu verba|For at least one human it's not true that: it's a child.}}
See that the '''na ku''' is placed before '''cu''', since a noun can go only before, not after the '''cu'''. Had I only used '''na''', it would have to go after '''cu''' — but that would have negated the entire clause, meaning "It's not true that: At least one human is a child".


By non-logical conjunctions, we mean that the truth of the combined terms does not depend on the truth of the individual components. It may not be true that '''la .kris. cu bevri lo pipno'''. ''Chris carries the piano'', or '''la .pat. cu bevri lo pipno''' ''Pat carries the piano'', for example (to revisit an example from [[#4|Lesson 4]]), even if it is true that '''la .kris. joi la .pat. cu bevri lo pipno''' ''Chris and Pat carry the piano.''
When the '''na ku''' is moved rightwards, any quantifier is inverted — that is: '''ro''' is turned into '''su'o'''. This is, of course, only if the meaning of the clause has to be preserved. This means that when the '''na ku''' is placed at the end of the clause, only the main verb is negated but all the nouns and prepositions are preserved, as can be seen by these three identical clauses:


Lojban has several other non-logical conjunctions; we'll cover the most frequently used ones:
{{mu|na ku ro lo verba cu tadni bu'u da poi ckule|It's not true that all children are students in a school.}}
*fa'u carries the meaning of ''respectively'': it relates pairs of sumti '''cross-wise'''. If I were to say
{{gl|ckule|x1 is a school at area x2}}


:la .alis. je la .jasmin. cu tavla la .djang. je la .ranjit.
{{mu|su'o lo verba cu tadni na ku bu'u da poi ckule|Some children study in not a single school.}}
{{mu|su'o lo verba cu tadni bu'u ro lo ckule na ku|Some children are for all schools don't study in them.}}


that means that both Alice and Jasmine talk to both Zhang and Ranjeet. If I want to say that Alice only talked to Zhang, and Jasmine only to Ranjeet (i.e. ''Alice and Jasmine talked to Zhang and Ranjeet, ''respectively''''), a logical conjunction is not useful. Instead, I would use fa'u to connect both pairs of sumti:'''
It is possible to use clause negation in all clauses, even the implicit clause of descriptive nouns. '''lo na prenu''' can refer to anything non-human, whether it be a sphinx, a baseball or the property of appropriateness.


:la .alis. fa'u la .jasmin. cu tavla la .djang. fa'u la .ranjit.
{{notci|Note: Like '''cu''', '''na''' can indicate that a main verb is coming up, so we can just say '''na''' instead of '''cu na'''. So you can say
:'''lo ninmu na nelci la .ian.''' {{=}} ''A woman likes Yan''
without adding '''cu''' before '''na nelci'''.}}


Alice, cross-wise with Jasmine, talks to Zhang, cross-wise with Ranjeet.
==== Interjections '''nai''', '''ru'e''', '''sai''', '''cu'i''' and '''na'i''' ====
#'''nai''' being an interjection attaches to the right of a construct and negates it.
#:{{mupli|'''tu bajra mlatu nai'''<br/>''That is a running non-cat.''}}
#:{{mupli|'''tu bajra nai mlatu'''<br/>''That is a non-running cat.''}}
#:As you can see '''nai''' is applied only to the last verb of a compound verb.
#When attempting to answer: “Is the king of the USA fat?”, all of these negations fail. While it's technically correct to negate it with na, since it makes no assumptions of that is true, it's mildly misleading since it could lead the listener to believe there is a king of the USA. For these scenarios, there is a metalinguistic negator, '''na'i'''.
#:'''na'i''' — metalinguistic negator. Something is wrong with assigning a truth value to the clause.
#:Because '''na'i''' has the grammar of interjections.
#:{{mupli|'''lo na'i pu te zukte be lo skami cu palci'''<br>''The sought goal (mistake!) of the computer was evil''.
#:'''palci''' = ''x1 is evil''
#:This sentence protests that computers can seek a goal volitionally.}}


*If you're talking about a range, you use '''bi'i''' to describe the range between the first thing and the second thing; so it corresponds to English ''between''. If you want to say ''I dropped my pencil somewhere between the office and the bar'', you would describe the location ''somewhere between the office and the bar'' as lo briju ku bi'i lo barja. The whole sentence would come out as:
==== Scalar negation ====
'''na'e''' attaches to the left of a construct and negates it implying the existence of a scale:
{{mu|tu na'e bajra mlatu|That is a other-than-running cat.}}
{{mu|tu bajra na'e mlatu|That is a running other-than-cat.}}
:(making us wonder what a non-cat might be)
:{{mupli|''My spouse is not a woman'' (meaning that he is a male).<br>'''lo speni be mi cu na'e ninmu''' or '''lo speni be mi cu to'e ninmu'''.
:Using scalar negation here implies that he exists, which '''na''' did not.}}


:mi falcru lemi pinsi ''vi lo briju bi'i lo barja''
Another unexpected effect compared to English ''not'':
:{{mupli|'''mi na nelci ro lo gerku''' means ''It is not true that I like each dog''.}} However,
:{{mupli|'''mi na'e nelci ro lo gerku''' means ''I don't like any dogs.''}}
While the mechanism of '''na ku''' resembles negation in English, it can be difficult to keep track of exactly what is negated and how that affects the clause. For that reason, the construct '''na ku''' is rarely seen anywhere other than the beginning of a clause. In most cases where more specific negation is needed people resort to a scalar negation using '''na'e'''. With '''na'e''' only a certain construcut is affected whereas with '''na ku''' everything to the right of it is negated. This is the reason '''na ku''' is used rather seldom and mostly at the beginning of the sentence. In
:{{mu|mi <u>na'e</u> nelci ro gerku|I other-than-like all dogs.}}
there is something that can be said about me and all dogs, but it's not that I like them.


{{notci|This selma'o, BIhI, like selma'o JOI to which all non-logical conjunctions belong, can join both sumti and selbri.}}
I might hate them:
{{mu|mi <u>to'e</u> nelci ro gerku|I anti-like ({{=}} dislike) all dogs.}}


*If the order of the things defining the range matters, you use '''bi'o'''. This corresponds to ''from... to...'' in English (though ''between'' covers both ordered and unordered intervals.) For example, ''from 1 PM to 2 PM'' is an interval lasting an hour; but ''from 2 PM to 1 PM'' would normally be interpreted as a 23-hour interval (1 pm the following day), since times in English are assumed to be presented in order. Lojban follows suit with li pavo lo'o bi'o li paci as a 23-hour interval. If I said li pavo lo'o bi'i li paci, the order of the two times would not matter at all; so I could still be talking about a one-hour interval instead.
'''to'e''' turns a main verb into its opposite: '''to'e nelci''' is pretty much the same thing as '''xebni''' ''to hate''.
 
Or I just might be neutral to them. I might write poems about them, or prescribe medicine for them, imitate them in polite company, or just might be indifferent to them. Then we'll use '''no'e''':
{{mu|mi <u>no'e</u> nelci ro gerku|I am neutral-as-to-liking all dogs.}}
But as for liking them, I don't.


{{notci|Tip: The selma'o BIhI needs ''all'' sumti terminated before it, not just normal sumti with lo or lo. Since numbers are also sumti, you have to use the terminator corresponding to li, which is lo'o.}}
Thus '''na'e''' is not a negator in the same sense as '''na'''. It doesn't state that a clause is false, but makes a positive statement that a clause is true – the same clause, but with a different main verb. This distinction is mostly academic, though. If, for example, I state '''mi na'e se nelci''' ''I am non-liked'', I actually state that some main verb applies to me, which is also on a relevant scale with the main verb '''nelci'''. Most of the time, we assume a scale where the positions are mutually exclusive like "like — dislike — hate", so '''mi na'e se nelci''' implies '''mi na se nelci'''. Therefore, the words '''no'e''' and '''to'e''' should only be used when the main verb is placed on some obvious scale: '''lo speni be mi cu to'e melbi''' – ''My spouse is ugly'' makes sense, since we immediately know what the opposite of beautiful is, while '''mi klama lo to'e zdani be mi''' – ''I go to my opposite thing of home'', while grammatical, leaves the listener guessing what the speaker's ''opposite of home'' is and should be avoided.
{{notci|Note: You can use non-logical conjunctions in forethought mode, too: the forethought conjunction is the non-logical conjunction followed by gi. So the forethought version of la .kris. joi la .pat. is joi gi la .kris. gi la .pat.}}
:{{mupli|''My spouse is not really a woman.''<br>'''lo speni be mi cu no'e ninmu'''}}
The scale here is presumed to be from woman to man.
:{{mupli|''I don't speak fast in English because I'm German.''<br>'''mi na'e sutra tavla bau lo glibau se ja'e lo nu mi dotco'''}}


{|
==== Chat ====
|-
<pre>
|The word "logical" in "logical connective" refers to the association a logical connective has with a truth function. Not all useful connectives can be defined through a truth function, however, and so there are other connectives beside the logical ones. <br>The meaning of a logical connective is defined the same as two different bridi connected with that logical connective. For instance, ''mi nitcu do .a la .djan.'' is defined to be equivalent to ''mi nitcu do .i ja mi nitcu la .djan.''. This definition is useful to bear in mind, because it implies that sometimes, sumti cannot be connected with logical connectives without chaning the meaning. Consider the sentence: "Jack and Joe wrote this play." One attempt at a translation would be: ''ti draci fi la .djak. e la .djous.'' <br>'''''draci''''' x1 is a drama/play about x2 by writer/dramatist x3 for audience x4 with actors x5
05:06 < nargavela> BTW I should ask on the mriste how to express the PEG '/' operator; it is a must for describing Lojban syntax in Lojban
|-
05:12 < nargavela> na broda = broda be naku
|The problem with this translation is that it means ''ti draci la .djak. ije ti draci la .djous.'', which is not really true. Neither Jack nor Joe wrote it, they did so together. What we want here is of course a mass, and some way to join Jack and Joe in one mass. This has little to do with a truth function so we must use a non-logical connective, which are of selma'o JOI. We'll return to this Jack and Joe-problem in a little - first: Four of the known JOI:
05:13 < markuja> .i lu na broda su'o da li'u na dunli lu broda be naku su'o da li'u
|}
</pre>


{|class="wikitable"
==== Negation and neutral meaning ====
|The Lojban connective:
'''nai''' is an interjection that makes a part of sentence negative in meaning. '''nai''' means ''no'' or ''not''.
|''ce''
*interjection in Lojban modifies the construct to the left of it.
|''ce'o''
{{mu|mi nelci nai do|I don't like you.|I like-not you (literally).}}
|''joi''
{{mu|mi nai nelci do|Not I like you.|(may be someone else likes you)}}
|''jo'u''
:So when put after certain part of the phrase like pronoun or a verb it modifies that verb.
|-
*if put in the beginning of a phrase interjection modifies the whole phrase:
|Joins sumti and forms a:
{{mu|nai mi nelci do|It's not true that I like you.}}
|set
:So in the beginning of the phrase '''nai''' negates the whole phrase.
|sequence
*we can put an interjection after different parts of the same phrase shifting the meaning.
|mass
|group of individuals
|}


{|border="0" cellspacing="0" cellpadding = "0" width="100%"
The interjection '''cu'i''' makes a part of sentence middle in its meaning.
|The functions of these words are simple: ''lo'i remna jo'u lo'i gerku'' considers both the set of
{{mu|mi nelci cu'i do|As for whether I love or hate you, I'm indifferent to you. I neither like nor hate you.}}


=== Raising ===
==== Double negation ====
:'''tu'a''' - convert noun to vague abstraction involving the noun. Equivalent to '''lo nu''' + the noun + '''co'e vau'''
Sometimes it's necessary to use double negation. Let's show how this works using examples from English and Chinese.


vau needed?
The structure 非...不可 (fēi … bùkě) is one of the most commonly used in Mandarin Chinese. It means "must"/"absolutely must"/"need to." 非 means "not"/"no" and 不可 means "not possible". It's literally translated as "not not possible."
:'''mi na cfifa'i ra na fau da'''
:我非批评她不可。
:''wǒ fēi pīpíng tā bùkě''
:<code>I can't not to criticize her</code>.
:''I absolutely must criticize her.''


So you can translate tu'a as ''some abstraction associated with...'', or more colloquially, ''some stuff about...''. tu'a is easily the most popular way of dealing with abstractions you wish weren't there in Lojban; Lojban sentences using it come out fairly similar to the natural language sentences ''without'' abstractions that we're used to seeing.
:'''mi'ai na tadni na fau da'''
:我们非学习不可。
:''wǒmen fēi xuéxí bùkě''
:''We must study.''


The other function of '''jai''' is easier to explain. It simply converts the selbri such that the noun in the x1 gets a '''tu'a''' in front of it ('''ko'a jai broda''' = '''tu'a ko'a broda'''). In other words, it converts the selbri in a way such that it builds an elliptical abstraction from the noun in the x1, and then fills x1 with the abstraction instead of the actual noun. Again, the original noun place is accessible by '''fai'''. A very active Lojban IRC-user often says '''le gerku pe do jai se stidi mi''', to use a random example of a noun in x1. What's he say?
:'''mi'ai tadni nai fau no da'''
:'''stidi''' = ''x1 inspires/suggests x2 in/to x3''
:<code>We can't not study.</code>
Answer:<span style="color: gray;background-color: gray; "> “I suggest (something about) your dog.</span>


When looking up words in the Lojban dictionary, you may have already noticed that, where languages like English have people or things as subjects and objects, Lojban often uses abstractions instead as gismu places. For example, in English, you say that ''someone'' is interesting, or ''something'' is interesting. In Lojban, you aren't really meant to say either. E.g.:
=== fau and if ===
:'''cinri''' - ''x1 (abstraction) interests/is interesting to x2<nowiki>; x</nowiki>2 is interested in x1''
There is some controversy about against using '''jo''' and '''ja nai''' for expressing ''if'' and ''only if''. First you shouldn't use '''ja nai''' (''if'') when '''jo''' (''only if'') is meant. And in natural languages ''if'' is strongly tied up with notions of causality, precondition, or deduction — none of which is particularly emphasized by '''ja nai''' and '''jo''' as they are strictly logical conjunctions. For example, '''ja nai''' will give a poor rendering of ''It's not true that, if I'm rich, I'm happy'' — which is decidedly not the same thing as ''It's not true that I'm either not rich or happy''! For that reason, you will see many Lojbanists avoiding '''jo''' and '''ja nai''' in ordinary speech, and instead using prepositions like '''fau''' - ''in the event of...'', '''va'o''' - ''under conditions...'', '''se ja'e''' - ''results from ... happening'', or '''ni'i''' - ''logically caused by...''.


In other words, as far as Lojban is concerned, it's not things or people that are interesting, but <u>actions</u> or <u>properties</u> involving those things or people. For example, Jasmine cannot be said to be interesting simply by virtue of being Jasmine; the way Lojban puts it, it's the things Jasmine does (or is) that are interesting — the way she talks about British sitcoms, her choice of headgear, her tendency to break into '80s songs after she's had a few drinks.
Better examples of using them in ordinary speech:


:'''fenki''' - ''x1 (action/event) is crazy/insane/mad/frantic/in a frenzy (one sense) by standard x2''
:'''ja gi nai je gi ro da poi badna zo'u da pelxu gi mi citka lo badna gi mi citka lo pelxu'''
People are called crazy. Only occasionally are actions also called crazy. In Lojban, however, the verb '''fenki''' describes events (or actions, which is the same). In other words, as far as Lojban is concerned, craziness lies in actions, not in people; a crazy person is by definition someone who does crazy actions.
:''If everything for every banana it is true that it is yellow, and I am eating a banana, then I am eating something yellow.''
:'''ja gi nai je gi ro pipybanfi cu crino gi la .kermit. cu pipybanfi gi la .kermit. cu crino'''
:''If every frog is green, and Kermit is a frog, then Kermit is green.''


{{notci|Note: This means that someone suffering from the particular forms of mental illness loosely called ‘crazy’ wouldn't be called '''fenki''' in Lojban — since their condition is not <u>primarily</u> a matter of socially unacceptable actions — but rather '''menli bilma''' - ''mentally ill''.}}
==== Non-logical conjunctions ====
:'''lo jisra joi lo djacu''' - ''the juice and the water, considered together''
This is the same as '''lo gunma be lo jisra je lo djacu'''.


What if you want to say ''Jasmine is interesting'', but you don't care or just unsure what verb to use? I should at least be able to say something like ''Jasmine {doing some stuff I'm not listing here} is interesting'', or ''Some things about Jasmine are interesting.'' In other words, I have to say
:'''djacu''' = ''x1 is some water''
:'''lo nu la .jasmin. cu co'e cu cinri'''
:'''jisra''' = ''x1 is some juice''
or
Not all English sentences containing ''and'' are like this, though. Firstly, sentences like ''I had a bath and washed my hair'' are structurally different and will be dealt with later on. Secondly, ''I visited Ranjeet and Jasmine'' is slightly different from ''I visited Ranjeet <u>and</u> I visited Jasmine.'' In this case, you probably want to say that you visited Ranjeet-and-Jasmine ''as a unit'' on one occasion — not that you visited Ranjeet and Jasmine on (potentially) different occasions (''It is true that I visited Ranjeet, and it is true that I visited Jasmine.'') In this case you don't want '''je''' (which is true but potentially misleading), but '''joi''', which means ‘in a mass with’. So what you have is
:'''tu'a la .jasmin. cu cinri'''
:'''mi pu vitke la .ranjit. joi la .jasmin.'''
:''I past visit Ranjeet in-a-mass-with Jasmine.''
:''I visited Ranjeet and Jasmine (together).''


Saying ''Zhang is crazy'' (or ''berserk'', probably a closer translation of '''fenki'''), I don't have to enumerate the various wacky stunts he has pulled over the years. I can simply say that ''some stuff about Zhang is crazy'', which in Lojban comes out as
This is just like the difference between '''lo ci gerku''' and '''loi ci gerku''' which we looked at in [[#4|Lesson 4]] — considering the three dogs as individuals, or as a mass. Incidentally, it is not just Lojban which makes this distinction; Turkish, for example, would use ''ile'' (‘with’) rather than ''ve'' (‘and’) for '''joi''' here.
:'''lo nu la .djang. cu <u>co'e</u> cu fenki'''


The value of co'e could be
By non-logical conjunctions, we mean that the truth of the combined terms does not depend on the truth of the individual components. It may not be true that '''la .kris. cu bevri lo pipno'''. ''Chris carries the piano'', or '''la .pat. cu bevri lo pipno''' ''Pat carries the piano'', for example (to revisit an example from [[#4|Lesson 4]]), even if it is true that '''la .kris. joi la .pat. cu bevri lo pipno''' ''Chris and Pat carry the piano.''
*'''dasni lo zirpu''' - ''wears purple''
*'''dansu la zgikrfanki jipci''' - ''dances the Funky Chicken''
*'''tavla bau la .lojban.''' - ''speaks Lojban''
or whatever; we're just not bothering to name it here.


So the usual Lojban for ''Jasmine is interesting'' is
Lojban has several other non-logical conjunctions; we'll cover the most frequently used ones:
*fa'u carries the meaning of ''respectively'': it relates pairs of sumti '''cross-wise'''. If I were to say


:'''<u>tu'a</u> la .jasmin. cu cinri'''
:la .alis. je la .jasmin. cu tavla la .djang. je la .ranjit.


and the usual Lojban for ''Kevin is crazy'' is
that means that both Alice and Jasmine talk to both Zhang and Ranjeet. If I want to say that Alice only talked to Zhang, and Jasmine only to Ranjeet (i.e. ''Alice and Jasmine talked to Zhang and Ranjeet, ''respectively''''), a logical conjunction is not useful. Instead, I would use fa'u to connect both pairs of sumti:'''


:'''<u>tu'a</u> la .kevin. cu fenki'''
:la .alis. fa'u la .jasmin. cu tavla la .djang. fa'u la .ranjit.


Lojbanists noticed how linguists have been analyzing these concepts in natural languages, and how they were coming up with their own versions of verbs. Often, what was a noun in one part of the sentence, and a verb in another part, were brought together and considered to be underlyingly part of the same abstraction noun.
Alice, cross-wise with Jasmine, talks to Zhang, cross-wise with Ranjeet.


A good example is the phrase ''I am difficult to annoy'' in English. At first sight, you might think that ''I'' is a noun of ''difficult''. And grammatically it is: it's the subject. But logically it isn't: what we're describing as difficult is not ''me''. We can't say:
*If you're talking about a range, you use '''bi'i''' to describe the range between the first thing and the second thing; so it corresponds to English ''between''. If you want to say ''I dropped my pencil somewhere between the office and the bar'', you would describe the location ''somewhere between the office and the bar'' as lo briju ku bi'i lo barja. The whole sentence would come out as:


*''Who is difficult?''
:mi falcru lemi pinsi ''vi lo briju bi'i lo barja''
*''Me (to annoy).''


What's actually going on is that, underlyingly, what is difficult is <u>to annoy me</u>: the action of getting me annoyed is what is hard to achieve — not me! This is why English also allows you to say ''It is difficult to annoy me'', and (if you squint a little) ''To annoy me is difficult''. And sure enough, Lojban expresses this concept according to that ‘underlying’ form:
{{notci|This selma'o, BIhI, like selma'o JOI to which all non-logical conjunctions belong, can join both sumti and selbri.}}


:lo nu fanza mi cu nandu
*If the order of the things defining the range matters, you use '''bi'o'''. This corresponds to ''from... to...'' in English (though ''between'' covers both ordered and unordered intervals.) For example, ''from 1 PM to 2 PM'' is an interval lasting an hour; but ''from 2 PM to 1 PM'' would normally be interpreted as a 23-hour interval (1 pm the following day), since times in English are assumed to be presented in order. Lojban follows suit with li pavo lo'o bi'o li paci as a 23-hour interval. If I said li pavo lo'o bi'i li paci, the order of the two times would not matter at all; so I could still be talking about a one-hour interval instead.
:The event of annoying me is difficult


So why did English pull that weird switcheroo with ''I am difficult to annoy''? Basically, because when we talk, we aren't concentrating in our minds on intangible abstractions like ''the event of annoying me'', let alone ''the state of Jasmine having certain unspecified properties.'' Instead, we run little stories in our head, with heroes and villains: concrete heroes and villains — people, for the most part. And as it happens, we make the subjects of our sentences be the heroes and villains we're concentrating on. (That's what a subject's ultimate job is: to present what we're concentrating on).
{{notci|Tip: The selma'o BIhI needs ''all'' sumti terminated before it, not just normal sumti with lo or lo. Since numbers are also sumti, you have to use the terminator corresponding to li, which is lo'o.}}
{{notci|Note: You can use non-logical conjunctions in forethought mode, too: the forethought conjunction is the non-logical conjunction followed by gi. So the forethought version of la .kris. joi la .pat. is joi gi la .kris. gi la .pat.}}


So by pulling a switcheroo like that, we're not talking about abstractions and events any more; the subject of the sentence is now our perennially favorite subject — namely ''me'': it's ''me'' that is difficult to annoy. (Yes, it ''is'' all about me...) This process is called in linguistics '''raising''', because it raises concrete subjects (and objects) we want to talk about, out of the haziness of an abstraction noun (or ‘clausal argument’, to use English logical terminology).
{|
|-
|The word "logical" in "logical connective" refers to the association a logical connective has with a truth function. Not all useful connectives can be defined through a truth function, however, and so there are other connectives beside the logical ones. <br>The meaning of a logical connective is defined the same as two different bridi connected with that logical connective. For instance, ''mi nitcu do .a la .djan.'' is defined to be equivalent to ''mi nitcu do .i ja mi nitcu la .djan.''. This definition is useful to bear in mind, because it implies that sometimes, sumti cannot be connected with logical connectives without chaning the meaning. Consider the sentence: "Jack and Joe wrote this play." One attempt at a translation would be: ''ti draci fi la .djak. e la .djous.'' <br>'''''draci''''' x1 is a drama/play about x2 by writer/dramatist x3 for audience x4 with actors x5
|-
|The problem with this translation is that it means ''ti draci la .djak. ije ti draci la .djous.'', which is not really true. Neither Jack nor Joe wrote it, they did so together. What we want here is of course a mass, and some way to join Jack and Joe in one mass. This has little to do with a truth function so we must use a non-logical connective, which are of selma'o JOI. We'll return to this Jack and Joe-problem in a little - first: Four of the known JOI:
|}


Another case is saying
{|class="wikitable"
:''Jasmine is interesting and beautiful.''
|The Lojban connective:
We could say
|''ce''
:'''tu'a la jasmin cu cinri .i la jasmin cu melbi'''
|''ce'o''
as '''melbi''' requires a noun as it's first (x1) place.
|''joi''
|''jo'u''
|-
|Joins sumti and forms a:
|set
|sequence
|mass
|group of individuals
|}


But that would be bulky. We have another method
{|border="0" cellspacing="0" cellpadding = "0" width="100%"
:'''tu'a la jasmin cu cinri''' = '''la jasmin cu jai cinri'''
|The functions of these words are simple: ''lo'i remna jo'u lo'i gerku'' considers both the set of


So '''jai''' extracts a noun from the verb place containing an event. Thus, it's very useful:
=== Raising ===
:'''la jasmin cu jai cinri vau je melbi'''
:'''tu'a''' - convert noun to vague abstraction involving the noun. Equivalent to '''lo nu''' + the noun + '''co'e vau'''
:''Jasmine is interesting and beautiful.''


Now both verbs have the same noun in common.
vau needed?


Another example. How do I say that someone is a cheat, or a deceiver? The verb for "to deceive" is:
So you can translate tu'a as ''some abstraction associated with...'', or more colloquially, ''some stuff about...''. tu'a is easily the most popular way of dealing with abstractions you wish weren't there in Lojban; Lojban sentences using it come out fairly similar to the natural language sentences ''without'' abstractions that we're used to seeing.
:'''tcica''' - ''x1 (event/experience) misleads/deceives/dupes/fools/cheats/tricks x2 into x3 (event/state)''


So '''lo tcica''' is a trick, not a trickster; a deception, and not a deceiver. To say that someone is a trickster or a deceiver, we need to use '''tu'a''': '''tu'a zo'e tcica'''. But you can't put '''lo''' in front of '''tu'a da''': the deceiver has to be the x1 of some verb, in order to get their own noun. So we say '''lo jai tcica'''.
The other function of '''jai''' is easier to explain. It simply converts the selbri such that the noun in the x1 gets a '''tu'a''' in front of it ('''ko'a jai broda''' = '''tu'a ko'a broda'''). In other words, it converts the selbri in a way such that it builds an elliptical abstraction from the noun in the x1, and then fills x1 with the abstraction instead of the actual noun. Again, the original noun place is accessible by '''fai'''. A very active Lojban IRC-user often says '''le gerku pe do jai se stidi mi''', to use a random example of a noun in x1. What's he say?
:'''stidi''' = ''x1 inspires/suggests x2 in/to x3''
Answer:<span style="color: gray;background-color: gray; "> “I suggest (something about) your dog.”</span>


'''jai''' changes the place structure of the verb involved. This works just like '''se''' changing the place structure of it's verb, swapping its first and second place. If we put '''jai''' in front of a verb, its x1 place changes from an event, to any noun contained <u>within</u> the event. Let's try this with a few sentences:
When looking up words in the Lojban dictionary, you may have already noticed that, where languages like English have people or things as subjects and objects, Lojban often uses abstractions instead as gismu places. For example, in English, you say that ''someone'' is interesting, or ''something'' is interesting. In Lojban, you aren't really meant to say either. E.g.:
*'''<u>lo nu la .kevin. cu dasni loi zirpu</u> cu fenki'''
:'''cinri''' - ''x1 (abstraction) interests/is interesting to x2<nowiki>; x</nowiki>2 is interested in x1''
*'''<u>la .kevin.</u> cu jai fenki'''


*'''<u>lo nu la .jasmin. cu co'e</u> cu cinri'''
In other words, as far as Lojban is concerned, it's not things or people that are interesting, but <u>actions</u> or <u>properties</u> involving those things or people. For example, Jasmine cannot be said to be interesting simply by virtue of being Jasmine; the way Lojban puts it, it's the things Jasmine does (or is) that are interesting — the way she talks about British sitcoms, her choice of headgear, her tendency to break into '80s songs after she's had a few drinks.
*'''<u>la .jasmin.</u> cu jai cinri'''


*'''<u>tu'a la .jasmin.</u> cu tcica la .alis.'''
:'''fenki''' - ''x1 (action/event) is crazy/insane/mad/frantic/in a frenzy (one sense) by standard x2''
*'''<u>la .jasmin.</u> cu jai tcica la .alis.'''
People are called crazy. Only occasionally are actions also called crazy. In Lojban, however, the verb '''fenki''' describes events (or actions, which is the same). In other words, as far as Lojban is concerned, craziness lies in actions, not in people; a crazy person is by definition someone who does crazy actions.


*'''<u>lo nu fanza mi</u> cu nandu'''
{{notci|Note: This means that someone suffering from the particular forms of mental illness loosely called ‘crazy’ wouldn't be called '''fenki''' in Lojban — since their condition is not <u>primarily</u> a matter of socially unacceptable actions — but rather '''menli bilma''' - ''mentally ill''.}}
*'''<u>mi</u> jai nandu'''


You'll notice that, with these new place structures, the Lojban phrases sound pretty much like their English equivalents. For example,
What if you want to say ''Jasmine is interesting'', but you don't care or just unsure what verb to use? I should at least be able to say something like ''Jasmine {doing some stuff I'm not listing here} is interesting'', or ''Some things about Jasmine are interesting.'' In other words, I have to say
 
:'''lo nu la .jasmin. cu co'e cu cinri'''
:'''la .jasmin. cu jai cinri'''
or
:''Jasmine is interesting.''
:'''tu'a la .jasmin. cu cinri'''


:'''la .jasmin. cu jai tcica la .alis.'''
Saying ''Zhang is crazy'' (or ''berserk'', probably a closer translation of '''fenki'''), I don't have to enumerate the various wacky stunts he has pulled over the years. I can simply say that ''some stuff about Zhang is crazy'', which in Lojban comes out as
:''Jasmine deceives Alice.''
:'''lo nu la .djang. cu <u>co'e</u> cu fenki'''


We can now do with '''jai''' those things we couldn't before. The Lojban for ''Jasmine is interesting and beautiful'', for example, is
The value of co'e could be
:'''la .jasmin. cu jai cinri vau je melbi'''
*'''dasni lo zirpu''' - ''wears purple''
That's because Jasmine goes in the x1 place of '''jai cinri''', just as it goes into the x1 place of '''melbi'''. And if I want to make a noun meaning ‘deceiver’ or ‘trickster’, I can use '''jai''' to do it:
*'''dansu la zgikrfanki jipci''' - ''dances the Funky Chicken''
:'''<u>tu'a la .jasmin.</u> cu tcica''' '''<u>la .jasmin.</u> cu jai tcica''' → '''lo jai tcica'''
*'''tavla bau la .lojban.''' - ''speaks Lojban''
or whatever; we're just not bothering to name it here.


However, '''mi jai nandu''' does <u>not</u> correspond to ''I am difficult to annoy.'' In switching a concrete noun for the original x1 — the event that was difficult — we have lost the abstraction itself: there is nothing in '''mi jai nandu''' that means ''to annoy''. But not to worry: Lojban allows you to keep the original event in the verb phrase by preceding it with '''fai'''. '''fai''' is a place tag like '''fa''' and '''fe'''. It effectively adds a new place to the verb phrase. So ''I am difficult to annoy'' is matched almost word-for-word by the Lojban sentence
So the usual Lojban for ''Jasmine is interesting'' is
:'''mi jai nandu fai lo ka fanza mi'''
And we can apply this pattern further afield; for example, ''the book took three months to write'' is in Lojban properly
:'''lo nu finti lo cukta cu masti li ci'''
:''<nowiki>To write the book had a month-duration of three.'</nowiki>''
???what is raising? Raising allows the slightly more familiar-looking
:'''lo cukta cu jai masti li ci fai lo ka se finti'''


'''jai''' allows you to talk about things in a way that is in many ways more natural.
:'''<u>tu'a</u> la .jasmin. cu cinri'''


=== Referring to main verb phrase ===
and the usual Lojban for ''Kevin is crazy'' is
I use BAI in such a way that whenever their source predicate has a noun place that could integrate the main bridi, then they will do so (I explained the same thing in the {to'e ri'a} thread).


{du'i} is one such BAI. I define it thus:
:'''<u>tu'a</u> la .kevin. cu fenki'''


broda du'i X ~= X dunli lo su'u broda
Lojbanists noticed how linguists have been analyzing these concepts in natural languages, and how they were coming up with their own versions of verbs. Often, what was a noun in one part of the sentence, and a verb in another part, were brought together and considered to be underlyingly part of the same abstraction noun.
 
A good example is the phrase ''I am difficult to annoy'' in English. At first sight, you might think that ''I'' is a noun of ''difficult''. And grammatically it is: it's the subject. But logically it isn't: what we're describing as difficult is not ''me''. We can't say:


So, I would say, for example:
*''Who is difficult?''
*''Me (to annoy).''


(1) do bajra du'i lo nu lo tirxu cu bajra
What's actually going on is that, underlyingly, what is difficult is <u>to annoy me</u>: the action of getting me annoyed is what is hard to achieve — not me! This is why English also allows you to say ''It is difficult to annoy me'', and (if you squint a little) ''To annoy me is difficult''. And sure enough, Lojban expresses this concept according to that ‘underlying’ form:
"Your run is like that of a tiger."


It is seldom the case that a concrete object is {dunli} to an abstraction, although it is not impossible. But usually, replacing the X in the above equation with a concrete object will create "non-sense". But we can use {tu'a X} when we don't want to specify the full abstraction, or when it's obvious.
:lo nu fanza mi cu nandu
:The event of annoying me is difficult


(2) do bajra du'i tu'a lo tirxu
So why did English pull that weird switcheroo with ''I am difficult to annoy''? Basically, because when we talk, we aren't concentrating in our minds on intangible abstractions like ''the event of annoying me'', let alone ''the state of Jasmine having certain unspecified properties.'' Instead, we run little stories in our head, with heroes and villains: concrete heroes and villains — people, for the most part. And as it happens, we make the subjects of our sentences be the heroes and villains we're concentrating on. (That's what a subject's ultimate job is: to present what we're concentrating on).
"You run like a tiger [runs]"


The expansion of this {tu'a X} is almost always {lo nu X no'a}, which corresponds to the English in []-brackets in (2).
So by pulling a switcheroo like that, we're not talking about abstractions and events any more; the subject of the sentence is now our perennially favorite subject — namely ''me'': it's ''me'' that is difficult to annoy. (Yes, it ''is'' all about me...) This process is called in linguistics '''raising''', because it raises concrete subjects (and objects) we want to talk about, out of the haziness of an abstraction noun (or ‘clausal argument’, to use English logical terminology).


.i tu'a zo du'i pu smuvanbi .i lo do kanla cu dunli lo nu lo lunra cu te gusni lo nicte kei lo ka ma kau ni melbi
Another case is saying
:''Jasmine is interesting and beautiful.''
We could say
:'''tu'a la jasmin cu cinri .i la jasmin cu melbi'''
as '''melbi''' requires a noun as it's first (x1) place.


=== Numbers ===
But that would be bulky. We have another method
:'''tu'a la jasmin cu cinri''' = '''la jasmin cu jai cinri'''


==== Trash ====
So '''jai''' extracts a noun from the verb place containing an event. Thus, it's very useful:
So the numbers from one to nine are as follows:
:'''la jasmin cu jai cinri vau je melbi'''
#pa
:''Jasmine is interesting and beautiful.''
#re
#ci
#vo
#mu
#xa
#ze
#bi
#so
Notice that the numbers repeat the vowels AEIOU. ''zero'' is '''no''' (think ''yes, we have no bananas'').


{| class="wikitable"
Now both verbs have the same noun in common.
|| pano
|| 10
|-
|| zebi
|| 78
|-
|| xanoci
|| 603
|-
|| vomusore
|| 4,592
|}


4,592 has a comma in it (or a full stop in some languages, just to make things confusing). We can't use a comma in Lojban, because that means ''separate these two syllables'' (as we saw in [[#1|Lesson 1]] with Lojbanized names like '''.zo,is.''' for ''Zoe''). What we say instead is '''ki'o'''. We don't ''have'' to use ki'o, but it can make things clearer. So 4,592 can also be read as vo ki'o musore. ki'o also has the advantage that if the following digits are all zeroes, we don't need to say them, so 3,000 is ci ki'o. You can remember ki'o easily if you think of ''kilo'' — a thousand. (The similarity is not coincidental).
Another example. How do I say that someone is a cheat, or a deceiver? The verb for "to deceive" is:
:'''tcica''' - ''x1 (event/experience) misleads/deceives/dupes/fools/cheats/tricks x2 into x3 (event/state)''


Just as we have a word for a comma, we also have one for a decimal point: pi. So 5.3 is mupici. In fact, pi is not always decimal; it's the point for whatever number base you're using. But that's a more advanced topic.
So '''lo tcica''' is a trick, not a trickster; a deception, and not a deceiver. To say that someone is a trickster or a deceiver, we need to use '''tu'a''': '''tu'a zo'e tcica'''. But you can't put '''lo''' in front of '''tu'a da''': the deceiver has to be the x1 of some verb, in order to get their own noun. So we say '''lo jai tcica'''.


{{notci|Tip: Don't get this mixed up with the number ''pi'' (π): 3.14159..., which has its own word in Lojban: '''pai''' — oddly enough.}}
'''jai''' changes the place structure of the verb involved. This works just like '''se''' changing the place structure of it's verb, swapping its first and second place. If we put '''jai''' in front of a verb, its x1 place changes from an event, to any noun contained <u>within</u> the event. Let's try this with a few sentences:
*'''<u>lo nu la .kevin. cu dasni loi zirpu</u> cu fenki'''
*'''<u>la .kevin.</u> cu jai fenki'''


When you want to talk about numbers as sumti — that is to say, as things in and of themselves — you need to put an article in front of them. But that article cannot be la, and for reasons which hopefully will become clear soon, it cannot be lo either. In front of numbers, Lojban uses the article li. So li pareci means ‘the number one hundred and twenty three’. ‘One, two, three’, on the other hand, would be li pa li re li ci: each li introduces a brand new number.
*'''<u>lo nu la .jasmin. cu co'e</u> cu cinri'''
*'''<u>la .jasmin.</u> cu jai cinri'''


=== Proportions ===
*'''<u>tu'a la .jasmin.</u> cu tcica la .alis.'''
:'''ci lo vo mensi pe mi cu nelci la .rikis.martin.'''
*'''<u>la .jasmin.</u> cu jai tcica la .alis.'''
:''Three of my four sisters like Ricky Martin.''


This states two facts:
*'''<u>lo nu fanza mi</u> cu nandu'''
#I have four sisters.
*'''<u>mi</u> jai nandu'''
#Three of them like Ricky Martin.
But it doesn't actually state that my fourth sister <u>hates</u> him — she may be indifferent to him, or never have heard of him.


One way out of this problem is to use '''fi'u''', which is like the Lojban slash sign. So ''two out of every three people'' is really ''2/3 of people'', or '''re fi'u ci lo prenu'''.
You'll notice that, with these new place structures, the Lojban phrases sound pretty much like their English equivalents. For example,


Of course, this is actually a fraction, and fractions have decimal equivalents. You could also say ''0.75'' - '''pi ze mu lo prenu'''.
:'''la .jasmin. cu jai cinri'''
:''Jasmine is interesting.''


WRONG! Yes, that's our new friend '''loi''' in that sentence. If I had said '''re fi'u ci lo prenu''', that would have to be understood in the same way as '''re lo prenu''' or '''ci lo prenu''' (i.e. as a count of individuals), and I would have ended up talking about two thirds of a person. In most cultures, chopping up persons into thirds is not considered acceptable behavior even for pollsters or advertisers. On the other hand, chopping up ''populations'' into thirds is perfectly acceptable; and that's what '''lo prenu''' is. (A population, I mean, not an acceptable. Though on second thought...)
:'''la .jasmin. cu jai tcica la .alis.'''
:''Jasmine deceives Alice.''


Here are some more proportions:
We can now do with '''jai''' those things we couldn't before. The Lojban for ''Jasmine is interesting and beautiful'', for example, is
:mi tcica pimu loi prenu
:'''la .jasmin. cu jai cinri vau je melbi'''
:I fooled half of the people (treating the people as a mass, or population)
That's because Jasmine goes in the x1 place of '''jai cinri''', just as it goes into the x1 place of '''melbi'''. And if I want to make a noun meaning ‘deceiver’ or ‘trickster’, I can use '''jai''' to do it:
:'''<u>tu'a la .jasmin.</u> cu tcica''' → '''<u>la .jasmin.</u> cu jai tcica''' → '''lo jai tcica'''


:mi tcica pafi'ure lei prenu
However, '''mi jai nandu''' does <u>not</u> correspond to ''I am difficult to annoy.'' In switching a concrete noun for the original x1 — the event that was difficult — we have lost the abstraction itself: there is nothing in '''mi jai nandu''' that means ''to annoy''. But not to worry: Lojban allows you to keep the original event in the verb phrase by preceding it with '''fai'''. '''fai''' is a place tag like '''fa''' and '''fe'''. It effectively adds a new place to the verb phrase. So ''I am difficult to annoy'' is matched almost word-for-word by the Lojban sentence
:I fooled one out of two people (which means exactly the same thing)
:'''mi jai nandu fai lo ka fanza mi'''
And we can apply this pattern further afield; for example, ''the book took three months to write'' is in Lojban properly
:'''lo nu finti lo cukta cu masti li ci'''
:''<nowiki>To write the book had a month-duration of three.'</nowiki>''
???what is raising? Raising allows the slightly more familiar-looking
:'''lo cukta cu jai masti li ci fai lo ka se finti'''
 
'''jai''' allows you to talk about things in a way that is in many ways more natural.
 
=== Referring to main verb phrase ===
I use BAI in such a way that whenever their source predicate has a noun place that could integrate the main bridi, then they will do so (I explained the same thing in the {to'e ri'a} thread).
 
{du'i} is one such BAI. I define it thus:


:mi tcica pa lo re mlatu
broda du'i X ~= X dunli lo su'u broda
:I fooled one out of the two cats (treating the cats as individuals)


:mi se slabu vopano lo pacivore gismu
So, I would say, for example:
:I am familiar with 410 out of the 1342 (existing) gismu
----
:mi na nelci ro lo gerku
:It's not true that I like all dogs.


(This is ''not'' the same as ''I don't like any dogs'', which would be '''mi nelci no lo gerku'''. There are other ways of saying this, but we haven't got enough grammar under our belt yet).
(1) do bajra du'i lo nu lo tirxu cu bajra
"Your run is like that of a tiger."


==== Summary<span id="4.6"></span> ====
It is seldom the case that a concrete object is {dunli} to an abstraction, although it is not impossible. But usually, replacing the X in the above equation with a concrete object will create "non-sense". But we can use {tu'a X} when we don't want to specify the full abstraction, or when it's obvious.


{| class="wikitable"
(2) do bajra du'i tu'a lo tirxu
|| la
"You run like a tiger [runs]"
|| that named
 
|-
The expansion of this {tu'a X} is almost always {lo nu X no'a}, which corresponds to the English in []-brackets in (2).
|| le{{^| what to do? }}
 
|| that one previously mentioned or known from context or this speech
.i tu'a zo du'i pu smuvanbi .i lo do kanla cu dunli lo nu lo lunra cu te gusni lo nicte kei lo ka ma kau ni melbi
|-
 
|| lo
=== Numbers===
|| that which is/does
So the numbers from one to nine are as follows:
|-
#pa
|| li
#re
|| the number
#ci
|}
#vo
#mu
#xa
#ze
#bi
#so
Notice that the numbers repeat the vowels AEIOU. ''zero'' is '''no''' (think ''yes, we have no bananas'').


('''lu''' is not an article, it's a quotation mark!)
{| class="wikitable"
{| class="wikitable"
|| la'e
|| pano
|| the referent of (not really an article, as it takes a full sumti or pro-sumti, as in la'edi'u, what the last sentence refers to, as opposed to di'u, the actual words of the last sentence).
|| 10
|-
|-
|| le'e
|| zebi
|| the typical previously mentioned or known from context
|| 78
|-
|-
|| lo'e
|| xanoci
|| the typical
|| 603
|}
{| class="wikitable"
|| lai
|| the mass named
|-
|-
|| lei
|| vomusore
|| the mass previously mentioned or known from context
|| 4,592
|-
|| loi
|| the mass which is/does
|}
{| class="wikitable"
|| la'i
|| the set named
|-
|| le'i
|| the set previously mentioned or known from context
|-
|| lo'i
|| the set which is/does
|}
|}


(Sets turn out to be pretty useful in Lojban, as we'll see towards the end of this course).
4,592 has a comma in it (or a full stop in some languages, just to make things confusing). We can't use a comma in Lojban, because that means ''separate these two syllables'' (as we saw in [[#1|Lesson 1]] with Lojbanized names like '''.zo,is.''' for ''Zoe''). What we say instead is '''ki'o'''. We don't ''have'' to use ki'o, but it can make things clearer. So 4,592 can also be read as vo ki'o musore. ki'o also has the advantage that if the following digits are all zeroes, we don't need to say them, so 3,000 is ci ki'o. You can remember ki'o easily if you think of ''kilo'' — a thousand. (The similarity is not coincidental).


We also looked briefly at '''lu'o''', which turns a set into a mass, and '''lu'a''', which turns a mass into a set of individuals (‘group’ and ‘ungroup’). Strictly speaking, these aren't articles, though.
Just as we have a word for a comma, we also have one for a decimal point: pi. So 5.3 is mupici. In fact, pi is not always decimal; it's the point for whatever number base you're using. But that's a more advanced topic.


=== More than ===
{{notci|Tip: Don't get this mixed up with the number ''pi'' (π): 3.14159..., which has its own word in Lojban: '''pai''' — oddly enough.}}


* mi poi zilkarbi do cu geltu
When you want to talk about numbers as sumti — that is to say, as things in and of themselves — you need to put an article in front of them. But that article cannot be la, and for reasons which hopefully will become clear soon, it cannot be lo either. In front of numbers, Lojban uses the article li. So li pareci means ‘the number one hundred and twenty three’. ‘One, two, three’, on the other hand, would be li pa li re li ci: each li introduces a brand new number.
* lo cabdei tcima poi zilkarbi lo prulamdei tcima cu glare
* 今天的天气比昨天的天气热。(Jin tian de tian qi bi zuo tian de tian qi re) = The weather today is hotter than the weather yesterday.
* 他的篮球比我的篮球好。(Ta de lan qiu bi wo de lan qiu hao) = He is better at basketball than me.


=== Build more precise words from basic ===
=== Proportions ===
Noun converters:
:'''ci lo vo mensi pe mi cu nelci la .rikis.martin.'''
*'''lo'i flora''' = ''bouquet''
:''Three of my four sisters like Ricky Martin.''
*'''lo'i taxfu''' = suit of clothes
*'''loi prenu''' = ''crowd''
Adverbs:
*'''ve'i cmana''' = ''hill''
*'''ve'u xamsi''' = ''ocean''
Suffixes:
* plisytricu = apple-tree
* plisygrute = apple (fruit)
-pre:
* vacrypre = airman
* brijypre = office worker
* ladrypre = milkman
* flalypre = lawyer
-stu - place:
*citkystu = dining-room
*vikmystu = toilet
*vitkystu =guestroom
*dracystu = drama theater
*lumcystu = bathroom
-sro - store:
* cidjysro = fridge, foodstore
-tci - instrument:
*katnytci = chisel
*ciskytci = typewriter, keyboard
*prinytci = printer


=== lujvo and Deep Gismu Structure ===
This states two facts:
'''jvajvo''' or regular compound words don't differ much from other verbs. The point in having jvajvo is to quickly create new meanings from existing ones.
#I have four sisters.
For example, if '''kalri''' means ''to be open'' then '''kalrygau''' means ''to open (something)''
#Three of them like Ricky Martin.
But it doesn't actually state that my fourth sister <u>hates</u> him — she may be indifferent to him, or never have heard of him.
 
One way out of this problem is to use '''fi'u''', which is like the Lojban slash sign. So ''two out of every three people'' is really ''2/3 of people'', or '''re fi'u ci lo prenu'''.


The rule is that we drop the last vowel from the gismu (root verb), add '''y''' and then add the suffix '''-gau'''.
Of course, this is actually a fraction, and fractions have decimal equivalents. You could also say ''0.75'' - '''pi ze mu lo prenu'''.


{{mupli|'''mi kalrygau lo canko''' {{=}} I open the window.
WRONG! Yes, that's our new friend '''loi''' in that sentence. If I had said '''re fi'u ci lo prenu''', that would have to be understood in the same way as '''re lo prenu''' or '''ci lo prenu''' (i.e. as a count of individuals), and I would have ended up talking about two thirds of a person. In most cultures, chopping up persons into thirds is not considered acceptable behavior even for pollsters or advertisers. On the other hand, chopping up ''populations'' into thirds is perfectly acceptable; and that's what '''lo prenu''' is. (A population, I mean, not an acceptable. Though on second thought...)


'''mi gasnu lo nu lo canko cu kalri'''
Here are some more proportions:
:mi tcica pimu loi prenu
:I fooled half of the people (treating the people as a mass, or population)


'''gau mi lo canko cu kalri'''
:mi tcica pafi'ure lei prenu
:I fooled one out of two people (which means exactly the same thing)


'''mi kargau lo canko'''}}
:mi tcica pa lo re mlatu
:I fooled one out of the two cats (treating the cats as individuals)


All four sentences mean the same. '''kargau''' is a synonym to '''kalrygau'''. What style you choose depends only on you. Thus you can safely speak without any jvajvo always resorting to prepositions like '''gau''' or full verbs like '''gasnu'''.
:mi se slabu vopano lo pacivore gismu
:I am familiar with 410 out of the 1342 (existing) gismu
----
:mi na nelci ro lo gerku
:It's not true that I like all dogs.


There is no need in memorizing '''kalrygau''' separately as it is easily derived from '''kalri''' once you know what the suffix '''-gau''' means.
(This is ''not'' the same as ''I don't like any dogs'', which would be '''mi nelci no lo gerku'''. There are other ways of saying this, but we haven't got enough grammar under our belt yet).


Here are some most important suffixes:
==== Summary<span id="4.6"></span> ====
*'''-gau''' means "to bring about, to be the agent of (some event)" just like we have the preposition '''gau'''. Thus compound verbs with '''-gau''' describe animated being performing actions like with ''me opening the window'' in the last example.
*'''-bi'o''' means "to turn"


*'''-se'e''' means "to cause (some event)" and describes events causing other events. It corresponds to the preposition '''seja'e''' and the verb '''se jalge'''.
{| class="wikitable"
*'''-dji''' means "to want", corresponds to '''djica'''
|| la
*'''-mau''' means "to exceed", corresponds to '''zmadu''', it creates comparative verbs. '''xamgu''' = ''good, '''xamgymau''' = ''better''.
|| that named
*'''-me'a''' means "to be less", corresponds to '''mleca''', it creates comparative verbs.
|-
*'''rai''' denotes extreme value. '''xamgyrai''' = ''best''.
|| le{{^| what to do? }}
|| that one previously mentioned or known from context or this speech
|-
|| lo
|| that which is/does
|-
|| li
|| the number
|}


We also have prefixes that are put to the beginning of verbs.
('''lu''' is not an article, it's a quotation mark!)
*'''nal-''' means ''not'' and corresponds to '''na'e'''
{| class="wikitable"
*'''tol-''' means ''anti-'' and corresponds to '''to'e'''.
|| la'e
*'''sel-''' corresponds to '''se''', '''tel''' or '''ter''' to '''te''', '''vel''' to '''ve''' and '''xel''' to '''xe'''.
|| the referent of (not really an article, as it takes a full sumti or pro-sumti, as in la'edi'u, what the last sentence refers to, as opposed to di'u, the actual words of the last sentence).
*:'''selycpacu''' = '''se spacu''' = ''to be received''
|-
*:'''terymukti''' = '''telymukti''' = '''te mukti''' = ''to intend to (do something)''
|| le'e
*'''zil-''' corresponds to '''zi'o''' and removes the first place from the verb
|| the typical previously mentioned or known from context
*'''mal-''' corresponds to '''mabla''' and gives a bad connotation of the verb
|-
*'''zan-''' corresponds to '''mabla''' and gives a bad connotation of the verb
|| lo'e
|| the typical
|}
{| class="wikitable"
|| lai
|| the mass named
|-
|| lei
|| the mass previously mentioned or known from context
|-
|| loi
|| the mass which is/does
|}
{| class="wikitable"
|| la'i
|| the set named
|-
|| le'i
|| the set previously mentioned or known from context
|-
|| lo'i
|| the set which is/does
|}


*'''nun-''' corresponds to '''nu''' and turns the verb into an event verb
(Sets turn out to be pretty useful in Lojban, as we'll see towards the end of this course).


If a verb starts with a vowel then you put a prefix, then '''<nowiki>y'</nowiki>''' and then the verb:
We also looked briefly at '''lu'o''', which turns a set into a mass, and '''lu'a''', which turns a mass into a set of individuals (‘group’ and ‘ungroup’). Strictly speaking, these aren't articles, though.
:'''akti''' = ''to be active, in operation''
:'''aktygau''' = ''to activate''
:'''toly'aktygau''' = ''to deactivate''


The '''y''' can be dropped if there is only one consonant from each side and both of them are either voiced or voiceless
=== Build more precise words from basic ===
==== Omitting '''y''' ====
Noun converters:
Sometimes you can omit '''y''' when creating a lujvo. The condition is the following:
*'''lo'i flora''' = ''bouquet''
#The consonant from each side of the '''y''' is not adjacent to other consonants
*'''lo'i taxfu''' = suit of clothes
#They are no the same consonant
*'''loi prenu''' = ''crowd''
#If one of the consonants is voiced ('''b,d,g,z,j,v''') and the other one is voiceless ('''p,t,k,s,c,f''') then '''y''' can't be omitted.
Adverbs:
Examples:
*'''ve'i cmana''' = ''hill''
#In '''kalrygau''' '''y''' can't be omitted as the rule 1 would be broken. There are two consonants to the left of '''y''': '''lr'''. However, in '''nunycasnu''' '''n''' to the left and '''c''' to the right are adjacent to vowels, not to consonants so we can remove '''y''' getting '''nuncasnu'''
*'''ve'u xamsi''' = ''ocean''
#
Suffixes:
 
* plisytricu = apple-tree
==== Negative lujvo ====
* plisygrute = apple (fruit)
{{notci|Note: '''na''' has its own rafsi, nar; but '''na'e''' is more useful in creating new words. na'e in a selbri still indicates an existing kind of relationship, which you would want to describe with a single lujvo<nowiki>; while </nowiki>na could mean anything, including non-existence — making it too broad a concept for most uses.}}
-pre:
 
* vacrypre = airman
{| class="wikitable"
* brijypre = office worker
|| lo naljmi
* ladrypre = milkman
|| one who does not understand (from jimpe ‘understand’)
* flalypre = lawyer
|-
-stu - place:
|| lo naljvi
*citkystu = dining-room
|| a non-competitor (from jivna ‘compete’)
*vikmystu = toilet
|-
*vitkystu =guestroom
|| lo nalkri
*dracystu = drama theater
|| a non-believer/skeptic (from krici ‘believe’)
*lumcystu = bathroom
|-
-sro - store:
|| lo nalyla'e
* cidjysro = fridge, foodstore
|| an unlikely event (from lakne ‘be likely’)
-tci - instrument:
|-
*katnytci = chisel
|| lo nalre'a
*ciskytci = typewriter, keyboard
|| a non-human (from remna ‘be human’)
*prinytci = printer
|}


We can see that nal is like the English ''non-'', but we need to remember that ''non-''sometimes has other meanings or associations that nal does not have. lo naljvi is simply someone who is not taking part in a competition, not a ‘non-contender’ in the sense of someone who competes but doesn't stand a chance of winning. Similarly lo nalre'a is someone who is not a member of the species ''homo sapiens'' (e.g. a chimpanzee or Klingon), and cannot be applied to someone who is inhumane or perceived as subhuman in some way.
=== lujvo and Deep Gismu Structure ===
'''jvajvo''' or regular compound words don't differ much from other verbs. The point in having jvajvo is to quickly create new meanings from existing ones.
For example, if '''kalri''' means ''to be open'' then '''kalrygau''' means ''to open (something)''
 
The rule is that we drop the last vowel from the gismu (root verb), add '''y''' and then add the suffix '''-gau'''.


We can also use nal with sel and its relatives; for example,
{{mupli|'''mi kalrygau lo canko''' {{=}} I open the window.
{| class="wikitable"
|| lo naltertcu
|| not a purpose/activity for which something is needed; something which has no requirements (from nitcu ‘x1<nowiki> needs/requires/is dependent on/[wants] necessity x</nowiki>2 for purpose/action/stage of process x3')
|-
|| lo nalveltu'i
|| an area of disagreement; a controversial issue (from tugni ‘x1<nowiki> [person] agrees with person(s)/position/side x</nowiki>2 that x3 (du'u) is true about matter x4')
|-
|| lo nalselzi'e
|| something you are not free to do (from zifre ‘x1 is free/at liberty to do/be x2 (event/state) under conditions x3')
|-
|| lo nalselsanji
|| something you are unaware of (from sanji ‘x1 is conscious/aware of x2 (object/abstract); x1 discerns/recognizes x2 (object/abstract)'; this gismu has no suitable short combining form)
|-
|| lo nalselse'i
|| someone who lacks a self/ego; an enlightened person according to Hindu/Buddhist philosophy (from sevzi ‘x1 is a self/ego/id/identity-image of x2')
|}


As you'll have guessed, the companions of na'e, namely to'e and no'e, have rafsi of their own: tol- and nor-, respectively. So ‘disinterested’, ‘uninterested’ and ‘bored’ in Lojban are norselci'i, nalselci'i and tolselci'i.
'''mi gasnu lo nu lo canko cu kalri'''


lujvo can be much more interesting than this; interesting enough, in fact, that we won't be covering them any further here. You can make lujvo out of pretty much any tanru you can devise; this is the main way to introduce ‘new words’ into Lojban. But to make the lujvo you come up with work, you need some background knowledge:
'''gau mi lo canko cu kalri'''


*how to make sure rafsi in a word stick together unambiguously in Lojban grammar (''The Complete Lojban Language'', Chapter 4.5–4.6, 4.10–4.12).
'''mi kargau lo canko'''}}
*how to make sure the gismu inside your tanru group together properly (''The Complete Lojban Language'', Chapter 5).
*how to derive the place structure of the lujvo from the place structures of the gismu that make it up (''The Complete Lojban Language'', Chapter 12).


It's worth your while to look into these issues if you'll be using the language seriously, and especially if you'll be writing in it. (lujvo are easier to deal with while writing than while speaking, because you have the time to reflect on how you'll be creating your new word.) At this stage, though, you don't need to go into all that just yet.
All four sentences mean the same. '''kargau''' is a synonym to '''kalrygau'''. What style you choose depends only on you. Thus you can safely speak without any jvajvo always resorting to prepositions like '''gau''' or full verbs like '''gasnu'''.


(See also: http://selpahi.weebly.com/17/post/2014/02/how-to-get-rid-of-formal-rafsi.html )
There is no need in memorizing '''kalrygau''' separately as it is easily derived from '''kalri''' once you know what the suffix '''-gau''' means.
 
All of the following affixes produce a regularly derivable brivla with at most one added place. Some will add zero places, some one place.
Here are some most important suffixes:
*'''-gau''' means "to bring about, to be the agent of (some event)" just like we have the preposition '''gau'''. Thus compound verbs with '''-gau''' describe animated being performing actions like with ''me opening the window'' in the last example.
(Note: The option to enter non-derivable definitions for words using a suffix should not be taken away entirely, as it would mean a restriction in what words we can make up. I would suggest to maintain at least a bit of freedom just in case. Still, in the vast majority of cases, the suffixes should yield easy to understand derivations.
*'''-bi'o''' means "to turn"
Extra note: In contrast, words that don't use any catalogued affixes are free to mean whatever their inventors choose. As long as it is morphologically a brivla, it can mean whatever. Lujvo/Gismu/Fu'ivla are no longer regarded as mutually exclusive groups, meaning that a "lujvo" shape can have a fu'ivla definition, or a fu'ivla can have gismu shape, etc. Another consequence is that type 3 fu'ivla and most kinds of type 4 fu'ivla have no reason to exist anymore. A word like {tcizbaga "cheeseburger" is a valid borrowing, because it's a brivla, and we no longer have to add annoying clusters to artificially change its type to fu'ivla.)
 
*'''-se'e''' means "to cause (some event)" and describes events causing other events. It corresponds to the preposition '''seja'e''' and the verb '''se jalge'''.
*'''-dji''' means "to want", corresponds to '''djica'''
Prefixes:
*'''-mau''' means "to exceed", corresponds to '''zmadu''', it creates comparative verbs. '''xamgu''' = ''good, '''xamgymau''' = ''better''.
*'''-me'a''' means "to be less", corresponds to '''mleca''', it creates comparative verbs.
These are always placed before a brivla.
*'''rai''' denotes extreme value. '''xamgyrai''' = ''best''.
 
Prefixes are preferably CCV- so that we don't have to worry about tosmabru. People are supposed to be able to use them spontaneously and creatively, and ideally tosmabru should not be a possible danger. CV[nlr]- is also tosmabru-safe.
We also have prefixes that are put to the beginning of verbs.
*'''nal-''' means ''not'' and corresponds to '''na'e'''
affix  source    effect        example
*'''tol-''' means ''anti-'' and corresponds to '''to'e'''.
---------------------------------------
*'''sel-''' corresponds to '''se''', '''tel''' or '''ter''' to '''te''', '''vel''' to '''ve''' and '''xel''' to '''xe'''.
cfa-  (cfari)    beginning    cfaviska "to catch sight of something"
*:'''selycpacu''' = '''se spacu''' = ''to be received''
zan-   (zabna)    positive conn. ...
*:'''terymukti''' = '''telymukti''' = '''te mukti''' = ''to intend to (do something)''
mal-  (mabla)    negative conn.
*'''zil-''' corresponds to '''zi'o''' and removes the first place from the verb
ran- ? (ranji)    continuously  ranbajra "run continuously" "keep running"
*'''mal-''' corresponds to '''mabla''' and gives a bad connotation of the verb
?      (purci)   ex-  
*'''zan-''' corresponds to '''mabla''' and gives a bad connotation of the verb
can-   (canlu)   around an area  canbajra "to run around"       
 
xru-   (xruti)  re/back  xrudunda "to give back"; xrudarxi "to hit back"
*'''nun-''' corresponds to '''nu''' and turns the verb into an event verb
pru-   (purci)   pre-/ahead   
 
...
If a verb starts with a vowel then you put a prefix, then '''<nowiki>y'</nowiki>''' and then the verb:
bra-  (barda)   big
:'''akti''' = ''to be active, in operation''
cma-   (cmalu)  small
:'''aktygau''' = ''to activate''
rem-   (remna)   human         
:'''toly'aktygau''' = ''to deactivate''
 
Suffixes:
The '''y''' can be dropped if there is only one consonant from each side and both of them are either voiced or voiceless
==== Omitting '''y''' ====
These are always placed at the end of a brivla.
Sometimes you can omit '''y''' when creating a lujvo. The condition is the following:
#The consonant from each side of the '''y''' is not adjacent to other consonants
affix  source    effect        example
#They are no the same consonant
---------------------------------------
#If one of the consonants is voiced ('''b,d,g,z,j,v''') and the other one is voiceless ('''p,t,k,s,c,f''') then '''y''' can't be omitted.
-bi    (binxo)    become        xunrybi "to become red" "to blush"
Examples:
-zu    (zukte)   action        crezenzu "to practice"
#In '''kalrygau''' '''y''' can't be omitted as the rule 1 would be broken. There are two consonants to the left of '''y''': '''lr'''. However, in '''nunycasnu''' '''n''' to the left and '''c''' to the right are adjacent to vowels, not to consonants so we can remove '''y''' getting '''nuncasnu'''
-mu    (mukti)    motive
#
-ri    (rinka)    cause        tatpyri "tiring"
 
-ja    (selja'e) result/outcome glekyja "result in being happy"
==== Negative lujvo ====
-gau  (gasnu)   bring about   jungau "tell"
{{notci|Note:  '''na''' has its own rafsi, nar; but '''na'e''' is more useful in creating new words. na'e in a selbri still indicates an existing kind of relationship, which you would want to describe with a single lujvo<nowiki>; while </nowiki>na could mean anything, including non-existence — making it too broad a concept for most uses.}}
-dji  (djica)    want          djunydji "to want to know"
 
-ju    (djuno)   know          klamyju "know that someone goes"
{| class="wikitable"
-tce  (mutce)    very       
|| lo naljmi
-ru    (ruble)
|| one who does not understand (from jimpe ‘understand’)
-mau  (zmadu)
|-
-me    (mleca)
|| lo naljvi
-rai  (traji)
|| a non-competitor (from jivna ‘compete’)
-ze    (zenba)
|-
-di    (jdika)    decrease      glarydi "to cool down"
|| lo nalkri
-cu/ki (cumki) possibility    cpacycu/cpacyki "available"/"acquirable"
|| a non-believer/skeptic (from krici ‘believe’)
-kre  (krefu)   re-/again    krezbasu "rebuild"
|-
-si/su/xu?  (simxu)    each other    pramysi "to love each other"
|| lo nalyla'e
-sa?  (simsa)    -oid/-like    remnysa "humanoid"
|| an unlikely event (from lakne ‘be likely’)
-lu?  (simlu)    seem          morsylu "to appear dead"
|-
-ni?  (ckaji jibni) -almost
|| lo nalre'a
-tu/ci?(tutci)    tool/utensil  ciskyci/ciskytu "writing utensil"
|| a non-human (from remna ‘be human’)
...
|}
-fe    (fetsi)   feminine      baknyfe
 
-na    (nakni)   masculine    baknyna
We can see that nal is like the English ''non-'', but we need to remember that ''non-''sometimes has other meanings or associations that nal does not have. lo naljvi is simply someone who is not taking part in a competition, not a ‘non-contender’ in the sense of someone who competes but doesn't stand a chance of winning. Similarly lo nalre'a is someone who is not a member of the species ''homo sapiens'' (e.g. a chimpanzee or Klingon), and cannot be applied to someone who is inhumane or perceived as subhuman in some way.
      ()        group/crowd  remny-CV "a crowd", cinfy-CV "pride of lions"
 
-sra  (srana)   -ly/pertaining to  cevnysra "godly"
We can also use nal with sel and its relatives; for example,
   
{| class="wikitable"
|| lo naltertcu
+ ranji, tutci, vasru, stuzi, tcaci, cmima, simsa, troci...; prenu??
|| not a purpose/activity for which something is needed; something which has no requirements (from nitcu ‘x1<nowiki> needs/requires/is dependent on/[wants] necessity x</nowiki>2 for purpose/action/stage of process x3')
...
|-
+ diminutive and augmentative suffixes
|| lo nalveltu'i
|| an area of disagreement; a controversial issue (from tugni ‘x1<nowiki> [person] agrees with person(s)/position/side x</nowiki>2 that x3 (du'u) is true about matter x4')
|-
|| lo nalselzi'e
|| something you are not free to do (from zifre ‘x1 is free/at liberty to do/be x2 (event/state) under conditions x3')
|-
|| lo nalselsanji
|| something you are unaware of (from sanji ‘x1 is conscious/aware of x2 (object/abstract); x1 discerns/recognizes x2 (object/abstract)'; this gismu has no suitable short combining form)
|-
|| lo nalselse'i
|| someone who lacks a self/ego; an enlightened person according to Hindu/Buddhist philosophy (from sevzi ‘x1 is a self/ego/id/identity-image of x2')
|}
 
As you'll have guessed, the companions of na'e, namely to'e and no'e, have rafsi of their own: tol- and nor-, respectively. So ‘disinterested’, ‘uninterested’ and ‘bored’ in Lojban are norselci'i, nalselci'i and tolselci'i.
 
lujvo can be much more interesting than this; interesting enough, in fact, that we won't be covering them any further here. You can make lujvo out of pretty much any tanru you can devise; this is the main way to introduce ‘new words’ into Lojban. But to make the lujvo you come up with work, you need some background knowledge:
 
*how to make sure rafsi in a word stick together unambiguously in Lojban grammar (''The Complete Lojban Language'', Chapter 4.5–4.6, 4.10–4.12).
*how to make sure the gismu inside your tanru group together properly (''The Complete Lojban Language'', Chapter 5).
*how to derive the place structure of the lujvo from the place structures of the gismu that make it up (''The Complete Lojban Language'', Chapter 12).
 
It's worth your while to look into these issues if you'll be using the language seriously, and especially if you'll be writing in it. (lujvo are easier to deal with while writing than while speaking, because you have the time to reflect on how you'll be creating your new word.) At this stage, though, you don't need to go into all that just yet.
 
(See also: http://selpahi.weebly.com/17/post/2014/02/how-to-get-rid-of-formal-rafsi.html )
   
   
Also consider "compound affixes" CVCV combining ideas from multiple other words to get a new affix, possible examples include "-bizu" from "-bi'ozu'e" and "-pazu" from "pagzu'e" (to participate).
All of the following affixes produce a regularly derivable brivla with at most one added place. Some will add zero places, some one place.
   
   
-sebi  (se bilga) obligatory    plejysebi "must be paid for" 
(Note: The option to enter non-derivable definitions for words using a suffix should not be taken away entirely, as it would mean a restriction in what words we can make up. I would suggest to maintain at least a bit of freedom just in case. Still, in the vast majority of cases, the suffixes should yield easy to understand derivations.
-secu/seni (seltcu/se nitcu) required tcidysecu/tcidyseni "required reading"
Extra note: In contrast, words that don't use any catalogued affixes are free to mean whatever their inventors choose. As long as it is morphologically a brivla, it can mean whatever. Lujvo/Gismu/Fu'ivla are no longer regarded as mutually exclusive groups, meaning that a "lujvo" shape can have a fu'ivla definition, or a fu'ivla can have gismu shape, etc. Another consequence is that type 3 fu'ivla and most kinds of type 4 fu'ivla have no reason to exist anymore. A word like {tcizbaga "cheeseburger" is a valid borrowing, because it's a brivla, and we no longer have to add annoying clusters to artificially change its type to fu'ivla.)
   
   
lo bi'u nai se cukta cu tcidysecu
"The book is required reading."
   
   
lo nu nerkla cu plejysecu
Prefixes:
"You have to pay to go in."
   
   
Another idea: A way to derive "easy to X" and "hard to X", e.g. "easy to provoke" or "hard to understand".
These are always placed before a brivla.
   
   
Prefixes are preferably CCV- so that we don't have to worry about tosmabru. People are supposed to be able to use them spontaneously and creatively, and ideally tosmabru should not be a possible danger. CV[nlr]- is also tosmabru-safe.
   
   
   
affix source    effect        example
Make a list of shapes that are safe from tosmabru no matter what affix gets added. Those word shapes are the most robust and should be preferred over the others.  
---------------------------------------
Safe:
cfa-  (cfari)    beginning    cfaviska "to catch sight of something"
CVC/CVCV
zan-  (zabna)    positive conn. ...
CCVCVCV
mal-  (mabla)    negative conn.
CCVCV
ran- ? (ranji)    continuously  ranbajra "run continuously" "keep running"
CVC/CV
?      (purci)    ex-   
can-  (canlu)    around an area  canbajra "to run around"       
xru-  (xruti)  re/back  xrudunda "to give back"; xrudarxi "to hit back"
pru-  (purci)  pre-/ahead   
...
...
bra-  (barda)  big
cma-  (cmalu)  small
rem-  (remna)  human         
   
   
Unsafe:
Suffixes:
CVCCV
...
These are always placed at the end of a brivla.
   
   
For safe zi'evla shapes, adding -CV suffixes is possible in two ways: Either by replacing the final vowel with -y- and adding -CV or by leaving the vowel alone and adding -CV. E.g. {delfinu} -> {delfinyfe}/{delfinufe}. Doing the latter with unsafe brivla shapes causes tosmabru. Doing it with CCVCV (safe) gismu shapes turns them into safe zi'evla shapes (e.g. gerkufe "female dog"), which could cause collisions with already existing zi'evla. Maybe what gets defined first wins.
affix  source    effect        example
 
---------------------------------------
=== zi'evla ===
-bi    (binxo)   become        xunrybi "to become red" "to blush"
''loan words'' are a way of introducing new words into Lojban, comparable to[[#8|lujvo]].
-zu    (zukte)   action        crezenzu "to practice"
 
-mu    (mukti)   motive
You got a brief taste of lujvo in [[#8|Lesson 8]]. As we said there, lujvo are the main way of introducing new words — more precisely, new brivla — into Lojban. The most important thing about lujvo is that, as selbri, they are meant to have very well-defined place structures; and there are guidelines in place for deriving them (see ''The Complete Lojban Language'', Chapter 12.) So, particularly when the concept you want to express is ‘verb-like’ (that is, when it's likely to have sumti of its own), lujvo are preferred.
-ri    (rinka)    cause        tatpyri "tiring"
 
-ja    (selja'e) result/outcome glekyja "result in being happy"
There are some cases, though, when you do have to borrow a word from another language, creating a loan word (called in Lojban a fu'ivla). This can be because the thing you're talking about is very concrete or particular, and/or because the reference is quite culture-specific. In either case, it would be really cumbersome to describe it with a combination of gismu. (For example, how would you come up with a description for ''brie''? Or ''rock 'n' roll''? — which, we should point out, you would have to keep distinct from the later musical genre of ''rock''!)
-gau  (gasnu)   bring about  jungau "tell"
 
-dji  (djica)   want         djunydji "to want to know"
The problem with borrowing words into Lojban is, Lojban has a quite thorough set-up for working out what the words are in a stream of letters. This means that most words you import into Lojban (once you spell them in Lojban letters) are likely to mean something else already. For example, if I want to bring the ancient name of the river Danube - ''Istros'' into Lojban, the last thing I want to do is start saying '''.istero'''. That will get analysed as '''.i stero''', which is something like ‘and steradian function’{{^| we need a better example }}.
-ju    (djuno)    know          klamyju "know that someone goes"
 
-tce  (mutce)    very       
The sanctioned way to deal with loan-words (described in more detail in ''The Complete Lojban Language'', Chapter 4.7) is to stick a gismu (minus its final letter) in front of the word, showing what sort of thing the word is; and to put an r (or, if an r is already there, an n) between the gismu and the word. The gismu helps the reader or listener, who has likely never seen this word before, guess what the word might be. This is particularly handy if the source word might be ambiguous between two different meanings. And the combination of gismu minus final vowel, source word (which should start with a consonant, and end with a vowel), and r or n will hopefully produce a cluster of consonants crunchy enough that it cannot be mistaken for another Lojban word or phrase.
-ru    (ruble)
 
-mau  (zmadu)
{{notci|Tip: There is no standard consonant to put in front of the word to become a fu'ivla if it starts with a vowel. Two popular choices are x and n. Similarly, there is no set convention on where to get the vowel from, if your word ends in a consonant. In these lessons, we'll just repeat the preceding vowel; e.g. ''England''  gugdr''n''ingl''a''nd''a'' (from gugde ‘country’).}}
-me    (mleca)
 
-rai  (traji)
So what does all this look like in practice? Well, we've already seen ''curry'':
-ze    (zenba)
 
-di    (jdika)    decrease      glarydi "to cool down"
*take ‘food’, <nowiki>cidj[a]</nowiki><nowiki>;</nowiki>
-cu/ki (cumki) possibility    cpacycu/cpacyki "available"/"acquirable"
*take the word in Lojban garb (starting with a consonant and ending with a vowel), kari<nowiki>;</nowiki>
-kre  (krefu)   re-/again    krezbasu "rebuild"
*and wedge them together with an r: cidjrkari.
-si/su/xu?  (simxu)    each other    pramysi "to love each other"
 
-sa?  (simsa)    -oid/-like   remnysa "humanoid"
(The consonant cluster is also crunchy enough to be difficult to pronounce; the r is a syllable on its own, and the word should sound something like ''shidgerrrrrkari'').
-lu?   (simlu)    seem          morsylu "to appear dead"
 
-ni?  (ckaji jibni) -almost
Loan words (in Lojban, fu'ivla) are still only sporadically used — particularly because, as of this writing at least, there is no Lojban dictionary where a standard list of them can be looked up. The problem of which language to borrow words from is also hard to settle, and the choices made can cause problems of their own. The most international solution for plant and animal names, for example, is Latin, and in particular the Latin of the Linnaean system of classification. But this means that, to come up with a word for ‘catnip’, say, you have to know Latin and your Linnaean taxonomy. (Or, like I did, look it up on the Internet — but you can't normally do that while you're having a conversation.) So fu'ivla are still largely unexplored terrain in Lojban.
-tu/ci?(tutci)   tool/utensil  ciskyci/ciskytu "writing utensil"
 
...
{{notci|Note: That said, you will occasionally see ‘Stage 4' fu'ivla in use. The fu'ivla we've seen are ‘Stage 3'; in Stage 4, you drop the initial ‘crunchy’ rafsi, reasoning that the word should already be well-known or recognisable enough — and making sure that the word still doesn't look like a normal brivla. (For example, ''The Complete Lojban Language'' suggests tci'ile for ‘Chile’, instead of gugdrtcile.) Not everyone likes them, so they're not yet all that common, and you'll usually get plenty of warning if someone is using them.
-fe    (fetsi)    feminine      baknyfe
 
-na    (nakni)   masculine    baknyna
P.S.: If you were wondering, by the way: cirlrbri, zgiknroknrolo, zgiknroko.}}
      ()         group/crowd  remny-CV "a crowd", cinfy-CV "pride of lions"
 
-sra  (srana)   -ly/pertaining to  cevnysra "godly"
=== Sets ===
   
If you are referring to an ''ordered set'' — a sequence of things, in other words — then to connect members of that set instead of '''je''' you use the conjunction '''ce'o'''. For example, the Lojban alphabet is a sequence, and you'd list it saying
:'''.abu ce'o by. ce'o cy. ce'o dy. ce'o .ebu'''...
+ ranji, tutci, vasru, stuzi, tcaci, cmima, simsa, troci...; prenu??
and so on. This is what '''liste''' (''list'') and '''porsi''' (''sequence'') expect as their first place.
...
 
+ diminutive and augmentative suffixes
=== Other types of anaphora ===
==== he and she ====
Also consider "compound affixes" CVCV combining ideas from multiple other words to get a new affix, possible examples include "-bizu" from "-bi'ozu'e" and "-pazu" from "pagzu'e" (to participate).
Several letters have a defined meaning, a type of thing that it refers to, reliably. By adding a suffix '''xoi''' to the right of them we get corresponding pronouns that refer back to the most recent living thing, the most recent person, object, abstract thing and many more. The letters are based on the first letter of some gismu to make them easier to remember. Below is a list of possible referents.
 
-sebi  (se bilga) obligatory    plejysebi "must be paid for" 
{| class="wikitable"
-secu/seni (seltcu/se nitcu) required tcidysecu/tcidyseni "required reading" 
|| fie
|| fetsi
lo bi'u nai se cukta cu tcidysecu
|| "she", female, feminine
"The book is required reading."
|-
|| nue
lo nu nerkla cu plejysecu
|| nakni
"You have to pay to go in."
|| "he", male, masculine
|-
Another idea: A way to derive "easy to X" and "hard to X", e.g. "easy to provoke" or "hard to understand".
|| pie
|| prenu
|| "he/she/'ey/they", person
|-
Make a list of shapes that are safe from tosmabru no matter what affix gets added. Those word shapes are the most robust and should be preferred over the others.
|| rie
Safe:
|| remna
CVC/CVCV
|| "he/she/'ey/they", human being
CCVCVCV
|-
CCVCV
|| jie
CVC/CV
|| jmive (gi'e na remna)
...
|| "it", animate, animal
|-
Unsafe:
|| due
CVCCV
|| dacti (gi'e na jmive)
...
|| "it", inanimate object
|-
For safe zi'evla shapes, adding -CV suffixes is possible in two ways: Either by replacing the final vowel with -y- and adding -CV or by leaving the vowel alone and adding -CV. E.g. {delfinu} -> {delfinyfe}/{delfinufe}. Doing the latter with unsafe brivla shapes causes tosmabru. Doing it with CCVCV (safe) gismu shapes turns them into safe zi'evla shapes (e.g. gerkufe "female dog"), which could cause collisions with already existing zi'evla. Maybe what gets defined first wins.
|| sue
|| stuzi
|| "it", place, location
|-
|| mie
|| mucti
|| "it", abstract, immaterial, thoughts, ideas
|}


Let us look at some of the amazing things we can do with this.
=== zi'evla ===
''loan words'' are a way of introducing new words into Lojban, comparable to[[#8|lujvo]].


:'''na go'i .i mi pu no roi zvati sue'''
You got a brief taste of lujvo in [[#8|Lesson 8]]. As we said there, lujvo are the main way of introducing new words — more precisely, new brivla — into Lojban. The most important thing about lujvo is that, as selbri, they are meant to have very well-defined place structures; and there are guidelines in place for deriving them (see ''The Complete Lojban Language'', Chapter 12.) So, particularly when the concept you want to express is ‘verb-like’ (that is, when it's likely to have sumti of its own), lujvo are preferred.
:''No. I have never been there''.


:'''fie dunda lo kargu junla nue'''
There are some cases, though, when you do have to borrow a word from another language, creating a loan word (called in Lojban a fu'ivla). This can be because the thing you're talking about is very concrete or particular, and/or because the reference is quite culture-specific. In either case, it would be really cumbersome to describe it with a combination of gismu. (For example, how would you come up with a description for ''brie''? Or ''rock 'n' roll''? — which, we should point out, you would have to keep distinct from the later musical genre of ''rock''!)
:''She gave him an expensive watch''.


:'''mi je lo mensi be mi pu klama sue .i ku'i fie na nelci sue'''
The problem with borrowing words into Lojban is, Lojban has a quite thorough set-up for working out what the words are in a stream of letters. This means that most words you import into Lojban (once you spell them in Lojban letters) are likely to mean something else already. For example, if I want to bring the ancient name of the river Danube - ''Istros'' into Lojban, the last thing I want to do is start saying '''.istero'''. That will get analysed as '''.i stero''', which is something like ‘and steradian function’{{^| we need a better example }}.
:''Me and my sister went there, but she didn't like it''.


:'''ko xrugau due lo pu te tolcri be fi do'''
The sanctioned way to deal with loan-words (described in more detail in ''The Complete Lojban Language'', Chapter 4.7) is to stick a gismu (minus its final letter) in front of the word, showing what sort of thing the word is; and to put an r (or, if an r is already there, an n) between the gismu and the word. The gismu helps the reader or listener, who has likely never seen this word before, guess what the word might be. This is particularly handy if the source word might be ambiguous between two different meanings. And the combination of gismu minus final vowel, source word (which should start with a consonant, and end with a vowel), and r or n will hopefully produce a cluster of consonants crunchy enough that it cannot be mistaken for another Lojban word or phrase.
:''Put it back where you found it''.


:'''mie cinri sidbo'''
{{notci|Tip: There is no standard consonant to put in front of the word to become a fu'ivla if it starts with a vowel. Two popular choices are x and n. Similarly, there is no set convention on where to get the vowel from, if your word ends in a consonant. In these lessons, we'll just repeat the preceding vowel; e.g. ''England''  gugdr''n''ingl''a''nd''a'' (from gugde ‘country’).}}
:''That is an interesting idea''.


==== rixire ====
So what does all this look like in practice? Well, we've already seen ''curry'':
To refer to the next-to-last noun, the third-from-last sumti, and so on, '''ri''' may be subscripted using the particle '''xi''':


:'''lo smuci .i lo forca .i la rik. cu pilno rixire .i la .alis. cu pilno riximu'''
*take ‘food’, <nowiki>cidj[a]</nowiki><nowiki>;</nowiki>
:''<nowiki>A spoon. A fork. Rick uses [repeat next-to-last]. Alice uses [repeat fifth-from-last].</nowiki>''
*take the word in Lojban garb (starting with a consonant and ending with a vowel), kari<nowiki>;</nowiki>
*and wedge them together with an r: cidjrkari.


Here '''rixire''', or '''ri'''-Number 2, skips '''la .rik.''' to reach '''lo forca'''. In the same way, '''riximu''', or '''ri'''-Number 5, skips '''la .alis.''', '''rixire''', '''la rik.''', and '''lo forca''' to reach '''lo smuci'''. As can clearly be seen, this procedure is barely practicable in writing, and would break down totally in speech.
(The consonant cluster is also crunchy enough to be difficult to pronounce; the r is a syllable on its own, and the word should sound something like ''shidgerrrrrkari'').
That's why we have '''ra'''.


==== ko'a ====
Loan words (in Lojban, fu'ivla) are still only sporadically used — particularly because, as of this writing at least, there is no Lojban dictionary where a standard list of them can be looked up. The problem of which language to borrow words from is also hard to settle, and the choices made can cause problems of their own. The most international solution for plant and animal names, for example, is Latin, and in particular the Latin of the Linnaean system of classification. But this means that, to come up with a word for ‘catnip’, say, you have to know Latin and your Linnaean taxonomy. (Or, like I did, look it up on the Internet — but you can't normally do that while you're having a conversation.) So fu'ivla are still largely unexplored terrain in Lojban.


Now there are plenty of KOhA sumti to go around. In fact, if you've run out of words by getting to ko'u, you can start over again with fo'a, fo'e ... fo'u. There is a problem, though: you have to remember (a) which sumti was assigned to which KOhA word, and (b) to assign the sumti in the first place. There's nothing to say that this will not become commonplace in future Lojban usage. Right now, however, there is a feeling that this is a little too calculated to work spontaneously. And Lojban cannot readily use the little hints natural languages pepper their grammar with (like gender and number), to keep track of who is who.
{{notci|Note: That said, you will occasionally see ‘Stage 4' fu'ivla in use. The fu'ivla we've seen are ‘Stage 3'; in Stage 4, you drop the initial ‘crunchy’ rafsi, reasoning that the word should already be well-known or recognisable enough — and making sure that the word still doesn't look like a normal brivla. (For example, ''The Complete Lojban Language'' suggests tci'ile for ‘Chile’, instead of gugdrtcile.) Not everyone likes them, so they're not yet all that common, and you'll usually get plenty of warning if someone is using them.


As a result, yet another strategy has been introduced to refer back to sumti. This strategy dates back from ‘Institute’ Loglan, before Lojban arose in its modern form. (Yes, Lojban has a history and a prehistory. No, we don't really have the time to go into them here.)
P.S.: If you were wondering, by the way: cirlrbri, zgiknroknrolo, zgiknroko.}}


==== Assigning pronouns<span id="7.2"></span> ====
=== Other types of anaphora ===
==== he and she ====
Several letters have a defined meaning, a type of thing that it refers to, reliably. By adding a suffix '''xoi''' to the right of them we get corresponding pronouns that refer back to the most recent living thing, the most recent person, object, abstract thing and many more. The letters are based on the first letter of some gismu to make them easier to remember. Below is a list of possible referents.


If we're telling a story in English, the meaning of, say, ''she'' keeps changing. At the moment, it means ‘Alice’, but if Alice's friend Jasmine walks into the bar, ''she'' could very well mean start meaning ‘Jasmine’. In Lojban, we can keep on using lo go'i, ri and their relatives, but there is an easier way of dealing with a larger cast of characters.
{| class="wikitable"
 
|| fie
What we do is assign pro-sumti as and when we need them, using the cmavo ''goi'' (which is like the Latin word ''sive'', or the English ''also known as (aka)''). The sumti assigned by goi are a series called KOhA, consisting of ko'a, ko'e, ko'i ... you get the idea?
|| fetsi
 
|| "she", female, feminine
{{notci|Note for lawyers (and frustrated non-lawyers): The equivalent in legal documents of goi is ''henceforth referred to as,'' and '''ko'a''' is something like ''the party of the first part.'' Lojban has in fact been proposed as the ideal language for law, where precision is of utmost importance. It would also allow non-lawyers to understand legal documents, which would be something of a miracle.}}
|-
 
|| nue
OK, let's go back to Alice's story. We start by saying
|| nakni
:la .alis. ''goi ko'a'' klama lo barja
|| "he", male, masculine
This means that from now on, every time we use ko'a, we mean ‘Alice’. The man she sees can then be ko'e, so we say
|-
:.i ko'a zgana lo nanmu goi ko'e
|| pie
Now every time we use ko'e, it means that particular man, so the full story so far reads:
|| prenu
:la .alis. goi ko'a klama lo barja .i ko'a ze'a pinxe loi vanju .i ko'a zgana lo nanmu goi ko'e .i ko'e melbi .i caku ko'e zgana ko'a
|| "he/she/'ey/they", person
(Note how the '''cu'''s have disappeared: '''ko'a''', like '''mi''', doesn't need them, since it can't join with a selbri to form a new selbri).
|-
|| rie
|| remna
|| "he/she/'ey/they", human being
|-
|| jie
|| jmive (gi'e na remna)
|| "it", animate, animal
|-
|| due
|| dacti (gi'e na jmive)
|| "it", inanimate object
|-
|| sue
|| stuzi
|| "it", place, location
|-
|| mie
|| mucti
|| "it", abstract, immaterial, thoughts, ideas
|}


Assigning '''ko'e''' to '''lo nanmu''' is actually better than starting the next sentence with lo nanmu. This is because lo nanmu simply means ''the thing I have in mind which I call ‘man’,'' which is not exactly the same as ''the man'' (it could, in theory, be something totally different). Some Lojbanists might even say that using lo like this is a bit malglico. (Or at least malrarbau ‘damned natural languages’: lots of languages have definite articles, and Lojban lo is no definite article).
Let us look at some of the amazing things we can do with this.


{{notci|Tip: If you combine '''ko'a/ko'e/ko'i/ko'o/ko'u''' with '''ri/ra/ru''', ''don't'' count '''ko'a'''-type pro-sumti when you're counting back. For example
:'''na go'i .i mi pu no roi zvati sue'''
:la .alis. cu rinsa ko'e .i ri cisma
:''No. I have never been there''.
doesn't mean that '''ko'e''' (the man, in this context) smiles, but that ''Alice'' smiles. Why? Because it is pointless to have a replacing word ('''anaphor'''), like ri, replace another replacing word, like ko'e. If you wanted the x1 of cisma to be ko'e, you would have simply said .i ko'e cisma, not .i ri cisma. It works out simpler to keep ri/ra/ru in reserve for more important things.}}


Let's continue by introducing Alice's friend Jasmine (if people are wondering where I get all these unusual names from, Jasmine is an old Gujarati friend of mine). We continue ....
:'''fie dunda lo kargu junla nue'''
:''She gave him an expensive watch''.


:la .jasmin. goi ko'i mo'ine'i klama .i ko'i rinsa ko'e
:'''mi je lo mensi be mi pu klama sue .i ku'i fie na nelci sue'''
:Jasmine (henceforth #3), goes into. #3 greets #2.
:''Me and my sister went there, but she didn't like it''.
:Jasmine comes in and says hello to the guy.


'''mo'ine'i''' is another space ‘tense’. mo'i indicates movement; ne'i means ‘inside’ (from the gismu, nenri). So mo'ine'i corresponds to the English preposition ''into'' (while ne'i on its own corresponds to ''inside'' or ''in''.) The way Lojban grammar works, mo'ine'i on its own is treated as mo'ine'i ku: a preposition with an omitted sumti. (Remember caku, which is exactly the same. Just as baku means ‘afterwards’ (relative to the here-and-now), mo'ine'i <nowiki>[ku]</nowiki> means something like ‘in(to)wards' — but is nowhere near as weird in Lojban as it is in English).
:'''ko xrugau due lo pu te tolcri be fi do'''
:''Put it back where you found it''.


mo'i is extremely useful, as it allows you to distinguish between location and motion. For example, ''I ran behind the bar'' in English is properly speaking ambiguous: are you running while behind the bar, or are you running with your final destination behind the bar? Lojban does not allow that ambiguity: '''mi bajra ti'a lo barja''' means the former, while '''mi bajra ''mo'i'' ti'a lo barja''' means the latter. In the example given above, ne'i klama would mean not that Jasmine comes in (from outside), but that she is going from somewhere to somewhere else, while inside. This kind of ambiguity may pass unnoticed by native English speakers, but speakers of languages which are more precise about direction find it extremely vague (Turkish, for example, has at least three words to translate ‘here’).
:'''mie cinri sidbo'''
:''That is an interesting idea''.


=== '''bo''' for linking sentences ===
==== rixire ====
To refer to the next-to-last noun, the third-from-last sumti, and so on, '''ri''' may be subscripted using the particle '''xi''':


''bo'' has another use, which seems separate from selbri grouping: It can also bind a sumtcita to an entire bridi, so that the content of the sumtcita is not a sumti, but the following bridi. This is best explained with an example. <br>'''''xebni''''' x1 hates x2 <br>''mi darxi do .i mu'i bo mi do xebni'' – "I hit you, with motivation that I hate you." Here the ''bo'' binds ''mu'i'' to the following bridi.<br>This is where the technical difference between tense sumtcita and other sumtcita is relevant. You see, when binding a normal sumtcita to a bridi with ''bo'', it means that the following bridi somehow fits into the sumti place of the sumtcita. For the reason of God Knows Why, binding one of the words ''ba'' or ''pu'' to a bridi has the exact opposite effect. For example, in the bridi ''mi darxi do .i ba bo do cinjikca'', one would assume that the second bridi is somehow filled into the sumti place of''ba'', meaning that the bridi first uttered took place in the future of the second bridi. That's not the case, however, and the correct translation of that utterance would be "I hit you. Afterwards, you flirt". This weird rule is actually one of the main obstacles to a unification of all sumtcita into one single word class. Another difference is that tense-sumtcita can be elided, even though they apply. This rule makes more sense, since we often can assume bridi is placed in a time and space, but we can't assume that the sumtcita of BAI applies.
:'''lo smuci .i lo forca .i la rik. cu pilno rixire .i la .alis. cu pilno riximu'''
=== More interjections ===
:''<nowiki>A spoon. A fork. Rick uses [repeat next-to-last]. Alice uses [repeat fifth-from-last].</nowiki>''
One final thing: if you want to know how someone feels about something, once again Lojban provides a fill-in-the-slot question word. The word asking the listener to fill in the interjection that best applies is '''pei'''. You can fill '''pei''' in with any interjection or '''sei'''-phrase. So if I ask you


'''pei''' can also explicitly ask for NAI or CAI alone, by following a UI cmavo. So a response to
Here '''rixire''', or '''ri'''-Number 2, skips '''la .rik.''' to reach '''lo forca'''. In the same way, '''riximu''', or '''ri'''-Number 5, skips '''la .alis.''', '''rixire''', '''la rik.''', and '''lo forca''' to reach '''lo smuci'''. As can clearly be seen, this procedure is barely practicable in writing, and would break down totally in speech.
That's why we have '''ra'''.


:.i .u'ipei do farlu lo pesxu
==== ko'a ====
:You fell into the mud! Funny, eh?


could well be ru'e: ''Kinda...'' Then again, it could also be naicai: ''Absolutely not, and I shall thank you never to mention it in my presence again.'' (Allowing for some latitude in translation..).
Now there are plenty of KOhA sumti to go around. In fact, if you've run out of words by getting to ko'u, you can start over again with fo'a, fo'e ... fo'u. There is a problem, though: you have to remember (a) which sumti was assigned to which KOhA word, and (b) to assign the sumti in the first place. There's nothing to say that this will not become commonplace in future Lojban usage. Right now, however, there is a feeling that this is a little too calculated to work spontaneously. And Lojban cannot readily use the little hints natural languages pepper their grammar with (like gender and number), to keep track of who is who.
{{notci|Note: selma'o UI4 specifies what ‘part’ of you is feeling the emotion — whether it is a physical, social, mental response, and so on. selma'o UI5 has some ‘left-over’ modifiers; we already saw in passing ga'i, which indicates haughtiness.}}


=== Time journeys ===
As a result, yet another strategy has been introduced to refer back to sumti. This strategy dates back from ‘Institute’ Loglan, before Lojban arose in its modern form. (Yes, Lojban has a history and a prehistory. No, we don't really have the time to go into them here.)
How can I say "broda from morning to evening"? I doubt {broda co'a lo cerni co'u lo vanci} would work?


<latro`a> while I would say that, it is indeed wrong, because it says "the event of brodaing ending, which happens at the evening, begins at the morning"
==== Assigning pronouns<span id="7.2"></span> ====


<latro`a> that is, the co'a tag scopes over the co'u tag, while you want them at the same scope level; as far as I know there's nothing that can be done about this with tag syntax
If we're telling a story in English, the meaning of, say, ''she'' keeps changing. At the moment, it means ‘Alice’, but if Alice's friend Jasmine walks into the bar, ''she'' could very well mean start meaning ‘Jasmine’. In Lojban, we can keep on using lo go'i, ri and their relatives, but there is an easier way of dealing with a larger cast of characters.


<latro`a> you can, however, say, {lo nu broda cu cfari ca lo cerni gi'e se fanmo ca lo vanci}
What we do is assign pro-sumti as and when we need them, using the cmavo ''goi'' (which is like the Latin word ''sive'', or the English ''also known as (aka)''). The sumti assigned by goi are a series called KOhA, consisting of ko'a, ko'e, ko'i ... you get the idea?


<danr> it might be a bit too formal for my purposes. I'm considering {broda fi'o cfari lo cerni fi'o fanmo lo vanci}
{{notci|Note for lawyers (and frustrated non-lawyers): The equivalent in legal documents of goi is ''henceforth referred to as,'' and '''ko'a''' is something like ''the party of the first part.'' Lojban has in fact been proposed as the ideal language for law, where precision is of utmost importance. It would also allow non-lawyers to understand legal documents, which would be something of a miracle.}}


<latro`a_> same problem. {fi'o} has the same scope phenomena
OK, let's go back to Alice's story. We start by saying
:la .alis. ''goi ko'a'' klama lo barja
This means that from now on, every time we use ko'a, we mean ‘Alice’. The man she sees can then be ko'e, so we say
:.i ko'a zgana lo nanmu goi ko'e
Now every time we use ko'e, it means that particular man, so the full story so far reads:
:la .alis. goi ko'a klama lo barja .i ko'a ze'a pinxe loi vanju .i ko'a zgana lo nanmu goi ko'e .i ko'e melbi .i caku ko'e zgana ko'a
(Note how the '''cu'''s have disappeared: '''ko'a''', like '''mi''', doesn't need them, since it can't join with a selbri to form a new selbri).


<tsani> you can use termsets to give terms equal scope
Assigning '''ko'e''' to '''lo nanmu''' is actually better than starting the next sentence with lo nanmu. This is because lo nanmu simply means ''the thing I have in mind which I call ‘man’,'' which is not exactly the same as ''the man'' (it could, in theory, be something totally different). Some Lojbanists might even say that using lo like this is a bit malglico. (Or at least malrarbau ‘damned natural languages’: lots of languages have definite articles, and Lojban lo is no definite article).


<tsani> {.i broda co'a ko'a ce'e co'u ko'e}
{{notci|Tip: If you combine '''ko'a/ko'e/ko'i/ko'o/ko'u''' with '''ri/ra/ru''', ''don't'' count '''ko'a'''-type pro-sumti when you're counting back. For example
:la .alis. cu rinsa ko'e .i ri cisma
doesn't mean that '''ko'e''' (the man, in this context) smiles, but that ''Alice'' smiles. Why? Because it is pointless to have a replacing word ('''anaphor'''), like ri, replace another replacing word, like ko'e. If you wanted the x1 of cisma to be ko'e, you would have simply said .i ko'e cisma, not .i ri cisma. It works out simpler to keep ri/ra/ru in reserve for more important things.}}


=== Scope ===
Let's continue by introducing Alice's friend Jasmine (if people are wondering where I get all these unusual names from, Jasmine is an old Gujarati friend of mine). We continue ....
13:41 < selpahi> {ze'i so'i roi plipe} vs {so'i roi ze'i plipe}, {ze'i lo mentu mi so'i roi [ri] cu plipe} vs {so'i roi ku mi plipe ze'i lo snidu}


{puku da renvi} = there were survivors.
:la .jasmin. goi ko'i mo'ine'i klama .i ko'i rinsa ko'e
:Jasmine (henceforth #3), goes into. #3 greets #2.
:Jasmine comes in and says hello to the guy.


{da puku renvi} = Some survived.
'''mo'ine'i''' is another space ‘tense’. mo'i indicates movement; ne'i means ‘inside’ (from the gismu, nenri). So mo'ine'i corresponds to the English preposition ''into'' (while ne'i on its own corresponds to ''inside'' or ''in''.) The way Lojban grammar works, mo'ine'i on its own is treated as mo'ine'i ku: a preposition with an omitted sumti. (Remember caku, which is exactly the same. Just as baku means ‘afterwards’ (relative to the here-and-now), mo'ine'i <nowiki>[ku]</nowiki> means something like ‘in(to)wards' — but is nowhere near as weird in Lojban as it is in English).


{mi co'u ponse da} ----> tolcfa fa lo nu li su'o kaidza lo ka mi ponse ke'a - i lost everything.
mo'i is extremely useful, as it allows you to distinguish between location and motion. For example, ''I ran behind the bar'' in English is properly speaking ambiguous: are you running while behind the bar, or are you running with your final destination behind the bar? Lojban does not allow that ambiguity: '''mi bajra ti'a lo barja''' means the former, while '''mi bajra ''mo'i'' ti'a lo barja''' means the latter. In the example given above, ne'i klama would mean not that Jasmine comes in (from outside), but that she is going from somewhere to somewhere else, while inside. This kind of ambiguity may pass unnoticed by native English speakers, but speakers of languages which are more precise about direction find it extremely vague (Turkish, for example, has at least three words to translate ‘here’).
{da co'u se ponse mi} - there is something i lost.


I wasn't sure whether you were saying you come here when you are writing, or you don't come here often because you are usually writing
=== '''bo''' for linking sentences ===


{mi ru'inai jundi mu'i lo nu finti} implies the former
''bo'' has another use, which seems separate from selbri grouping: It can also bind a sumtcita to an entire bridi, so that the content of the sumtcita is not a sumti, but the following bridi. This is best explained with an example. <br>'''''xebni''''' x1 hates x2 <br>''mi darxi do .i mu'i bo mi do xebni'' – "I hit you, with motivation that I hate you." Here the ''bo'' binds ''mu'i'' to the following bridi.<br>This is where the technical difference between tense sumtcita and other sumtcita is relevant. You see, when binding a normal sumtcita to a bridi with ''bo'', it means that the following bridi somehow fits into the sumti place of the sumtcita. For the reason of God Knows Why, binding one of the words ''ba'' or ''pu'' to a bridi has the exact opposite effect. For example, in the bridi ''mi darxi do .i ba bo do cinjikca'', one would assume that the second bridi is somehow filled into the sumti place of''ba'', meaning that the bridi first uttered took place in the future of the second bridi. That's not the case, however, and the correct translation of that utterance would be "I hit you. Afterwards, you flirt". This weird rule is actually one of the main obstacles to a unification of all sumtcita into one single word class. Another difference is that tense-sumtcita can be elided, even though they apply. This rule makes more sense, since we often can assume bridi is placed in a time and space, but we can't assume that the sumtcita of BAI applies.
----
=== More interjections ===
"i come here rarely in order to write" - "in order to write i come here rarely"
One final thing: if you want to know how someone feels about something, once again Lojban provides a fill-in-the-slot question word. The word asking the listener to fill in the interjection that best applies is '''pei'''. You can fill '''pei''' in with any interjection or '''sei'''-phrase. So if I ask you
----
{mi na klama mu'i tu'a lo carvi} (“It is not true that [I go motivated by the rain]”), when they actually mean {mi na'e klama mu'i tu'a lo carvi} (“[I don't go], which is motivated by the rain”).  


< durka42> gleki: the classic example of going to the bank in the rain, for one
'''pei''' can also explicitly ask for NAI or CAI alone, by following a UI cmavo. So a response to
10:30 < gleki> durka42: pardon?
10:31 < lai_az> le nu mutce nandu mi cu rinka zo'e bi'unai
10:31 < lai_az> or zo'e bi'u if its new
10:31 < lai_az> or ti etc
10:31 < durka42> gleki: mi na klama lo banxa ki'u lo nu carvi
10:32 < gleki> sounds like "i dont drink alcohol for religious reasons"
10:32 < durka42> same, yes
10:33 < gleki> but how to translate it? for me the straightforward translations is a mental puzzle
10:33 < durka42> na-ki'u: religion isn't the reason I drink alcohol; ki'u-na: due to religion I don't drink
10:46 < gleki> and with the rain?
10:46 < durka42> na-ki'u: rain doesn't draw me to the bank, or something, u'i


Following Lojban usage, I observe an extremely high number of scope mistakes. It appears that people are generally unaware. Their sentences are usually still understood "thanks" to the human capability of letting context fix scope dependencies. That's why these mistakes tend to go unnoticed. Nobody corrects these mistakes, and they keep on being made. It's a problem, but it's not too late to get a new awareness of scope.One of the most common scope mistake is this:
:.i .u'ipei do farlu lo pesxu
:You fell into the mud! Funny, eh?


(1) mi na terve'u lo stagi ki'u lo nu mi na bevri lo jdini
could well be ru'e: ''Kinda...'' Then again, it could also be naicai: ''Absolutely not, and I shall thank you never to mention it in my presence again.'' (Allowing for some latitude in translation..).
"I did not buy vegetables because I didn't have money with me."
{{notci|Note: selma'o UI4 specifies what ‘part’ of you is feeling the emotion — whether it is a physical, social, mental response, and so on. selma'o UI5 has some ‘left-over’ modifiers; we already saw in passing ga'i, which indicates haughtiness.}}


If you just read the English, your mind is likely to autocorrect the scope, causing you to miss the problem in the Lojban sentence. Though this might seem like an obvious mistake, I just saw someone make it again today.
=== Time journeys ===
How can I say "broda from morning to evening"? I doubt {broda co'a lo cerni co'u lo vanci} would work?


Sentence (1) actually means:
<latro`a> while I would say that, it is indeed wrong, because it says "the event of brodaing ending, which happens at the evening, begins at the morning"
"That I bought vegetables because I didn't have money with me is false."


That's how the Lojban should be understood, but it often isn't. A practical fix is to use {.i ki'u bo}:
<latro`a> that is, the co'a tag scopes over the co'u tag, while you want them at the same scope level; as far as I know there's nothing that can be done about this with tag syntax


(2) mi na terve'u lo stagi .i ki'u bo mi na bevri lo jdini
<latro`a> you can, however, say, {lo nu broda cu cfari ca lo cerni gi'e se fanmo ca lo vanci}
"I didn't buy vegetables. The reason for that is that I didn't have money with me."


The other method requires forethought, which might seem annoying, but you'll soon come to realize (if you haven't already) that forethought is needed in all of Lojban the moment you begin to scope-juggle.
<danr> it might be a bit too formal for my purposes. I'm considering {broda fi'o cfari lo cerni fi'o fanmo lo vanci}


(3) mi ki'u lo nu na bevri lo jdini cu na terve'u lo stagi
<latro`a_> same problem. {fi'o} has the same scope phenomena
"I, because [I] didn't have money with me, did not buy vegetables."


Case in point?
<tsani> you can use termsets to give terms equal scope


There are also cases of scope mistakes that I observe even among otherwise experienced Lojbanists. This is usually the case when multiple tenses are being used. Often, people just tack on an additional tense at the end of their already-tensed phrase, and the result is usually a sentence with incorrect scope, which the speaker doesn't notice.
<tsani> {.i broda co'a ko'a ce'e co'u ko'e}


There is a difference between {ca pu} and {pu ca}. There is a difference between {pu co'a} and {co'a pu}. You normally cannot reverse the order of phrase operators and expect to get the same result each time. While initially a difficulty, it's also one of the strong points of Lojban. It's the very thing Lojban is supposed to be good at. To neglect this aspect is to neglect Lojban.
=== Vocatives ===
*When it isn't your turn to speak, but you want to barge in anyway, you can say '''ta'a''' — though it probably won't make anyone any happier that you're interrupting.
*Finally, to close communication (radio's ''Over and out!''), you can use '''fe'o'''. (This is what people actually <u>should</u> be putting at the end of their e-mails; but it's not as well-known a word as '''co'o''')


Learning to use this feature correctly requires one to rethink. Lojban is not English. It's not even close. It just happens to sometimes be able to mirror the word order. It quickly stops doing that once the statements get more complex, at which time it drifts more towards SOV languages. It's not unnatural for a language to require constant forethoughts. Japanese and Turkish do it, just to name two. (Also note that when I say "forethought", I'm not talking about forethought connectives, although they can help. I'm talking about the general concept of thinking ahead).


Simply put, you have to learn (practically by rote, or by making the same mistake many times and correcting it until it sticks) what operators should precede what operators in what situation. Having some mental categories of different situations is very helpful and, I assume, inevitable. Well, to be sure, there are always other methods, but this one doesn't seem so painful. To give you an example, you need to simply know that a prenexed quantifier is required in this general instruction manual sentence:
=== mekso saske ===
Likewise, you can attach a sumtcita: le nu darxi kei be gau do: “The event of hitting, which is caused by you”. Note that the presence or absence of kei makes it parse differently: With the fa'orma'o present, be attaches to nu, without the fa'orma'o, it attaches to darxi. So it decides what is being emphasised: Is the hitting, or the event of hitting caused by you? In this specific case, though, that's just about the same thing.


(4) "Thirdly, everyone holds on to the rope."
famyma'o Warning!: One could have description nouns inside description nouns, saying '''le le se cinjikca be mi ku gerku, = le gerku pe le se cinjikca be mi =“the dog of the man I'm flirting with”, but that's not very easy to read (or to understand when spoken), and is often being avoided.'''


Suppose this is a sentence that describes a hypothetical setup that the instruction manual is trying to explain to you. We can argue that the rope is not definite. It can be any rope (so {lo} fits more than {le}), but everyone is supposed to hold on to the same rope, that's the important part. The naive translation fails:
One need also to learn tu'a, since it will make a lot of sentences much easier. It takes a sumti and converts it to another sumti - an elliptical abstraction which has something to do with the first sumti. For example, I could saymi djica lo nu mi citka lo plise, or I could let it be up to context what abstraction about the apple I desire and just say mi djica tu'a lo plise. One always has to guess what abstraction the speaker means by tu'a SUMTI, so it should only be used when context makes it easy to guess. Another example: gasnu “x1 does/brings about x2 (volition not implied)” za'a do gasnu tu'a lo skami - “I see that you make the computer do something”. Officially, tu'a SUMTI is equivalent to le su'u SUMTI co'e. Vague, but useful. There are situations where you cannot use tu'a, even though it would seem suitable. These situations are when I don't want the resulting sumti to be an abstraction, but a concrete sumti. In this case, one can use zo'e pe. tu'a convert sumti to vague abstraction involving the sumti. Equivalent to le su'u SUMTI co'e kei ku


(4a) ci mai .ei ro lo prenu cu jgari lo skori
Finally, one kind of sumti can be turned into another by the words of the class LAhE. lu'a -Convert individual(s)/mass/sequence/set to individuals. lu'i - Convert individual(s)/mass/sequence/set to a set. lu'o - Convert individual(s)/mass/sequence/set to mass. vu'i - Convert individual(s)/mass/sequence/set to sequence; the order is not stated. The use of these words is straight-forwardly: Placing them before a sumti of a certain type makes a new sumti of a new type. Notice, though, that as a fourth kind of sumti, a sequence has been introduced. This is not used very often (it doesn't have its own gadri, for instance), but just included here for completion. The last two members of LAhE do not convert between types of sumti, but allows you to speak of a a sumti by only mentioning something which references to it: If one sumti A refers to a sumti B, for instance because sumti A is a title of a book, or a name, or a sentence (which always refer to something, at least a bridi), ‘'la'e SUMTI A refers to sumti B. For instance, ‘'mi nelci la'e di'ufor “I like what you just said” (not mi nelci di'u which just means "I like your previous sentence") or ‘'la'e le cmalu noltru for the book “The Little Prince”, and not some little prince himself. The cmavo ‘'lu'e does the exact reverse – ‘'lu'e SUMTI'' refers to an object which refers to the sumti. la'e - “the thing referred to by” - extracts a sumti A from a sumti B which refers to A. lu'e - “the thing referring to” - extracts a sumti B from a sumti A, when B refers to A.''


This doesn't tell us that everyone is supposed to hold on to the same rope. And we also cannot use {le} (and even if we could, it's not really the point of this example). (4a) is vague about the rope business. So what now? As I said, this type of situation needs a prenexed quantifier:
----
In order to clarify that all tenses are relative to the speaker's current position, the word'''nau''' can be used at any time. Another word, '''ki''' marks a tense which is then considered the new standard. That will be taught way later. '''nau''' updates temporal and spacial frame of reference to the speaker's current 'here and now'.
----
In the Lojban tense system, all tenses are prepositions, which we have conveniently just made ourselves familiar with. Okay okay, technically, tenses are slightly different from other prepositions, but this difference is almost insignificant, and won't be explained until later. In most aspects they are like all other prepositions; they are terminated by '''ku''', making it explicit that selma'o (the group) PU is terminated by '''ku'''.


(4b) ci mai .ei pa da poi skori zo'u: ro lo prenu da jgari
*Every time '''nu''' starts an abstraction — a phrase nested inside another phrase —'''kei''' ends it.
*Every time an article like '''lo''' or '''loi''' starts a sumti, '''ku''' ends it.
*Every time a string of numbers starts, '''boi''' ends it.
*Every time a series of sumti follows a selbri, '''vau''' ends it.
*The <u>binding</u> can be terminated with be'o - one instance of be'o for each selbri which has sumti bound by be. To list them: be binds sumti or sumtcita to selbri bei binds a second, third, fourth (etc.) sumti or sumtcita to a selbri be'o ends binding to selbri
*li'u is one of the few terminators that can almost never be missed out, since that would make everything else that follows part of the quotation.  
*The elidable terminator for '''sei''' is “se'u” (of selma'o SEhU); it is rarely needed, except to separate a selbri within the '''sei''' comment from an immediately following selbri (or component) outside the comment.


This tells us explicitly that each person is supposed to hold on to the same rope.
For example, if we see '''cu''' in a sentence, we know that what is coming up is a selbri<nowiki>; so the </nowiki>sumti before it must now be over. So we can drop the'''ku'''. (In fact, that's why '''cu''' exists in the first place: the beginning of a verb is a much more important structural break in natural languages than the end of a noun.) If a new sentence is beginning — as signalled by perhaps the most distinctively Lojbanic word, the ‘audible punctuation’ '''.i''' — then (have we covered .i yet?) there can be no more sumti from the old sentence; so we drop the '''vau'''. In fact, it is only in situations of potential ambiguity, like the sentence we've been looking at, that you'll get terminators appearing in normal Lojban usage at all.  


Yes, you can also just reverse the order of the sumti as in (5), but it would be a different kind of word order (which proves my point, but messes up the example. If you want to mirror the English, (4b) is the way to go).
{{notci|Note: Remember those pesky possessive constructions from [[#3|Lesson 3]], when you couldn't flip '''lo tamne pe lo ninmu klama''' the other way around, because it was ambiguous? All you need is '''ku''' to resolve that ambiguity: '''lo lo ninmu klama ku tamne''' means 'the woman traveller's cousin', and lo lo ninmu ku klama tamne means ‘the woman's traveller cousin.


(5) ci mai .ei pa skori cu se jgari ro lo prenu
Still, most Lojbanists think the flip-around is not worth the hassle of inserting that bothersome '''ku''', so you rarely see it used when the ‘possessor’ sumti is not a one-word sumti.}}(ping!)


This one is shorter, not requiring an explicit prenex, and no explicit {da} binding. It does require one to change the order of the arguments, though, which might not always be desirable. Sometimes, placing a sumti in the x1 achieves a certain effect, e.g. topicalization and switching arguments can potentially ruin that.
=== cmene ===
The final point is stress. As we've seen, Lojban words are stressed on the penultimate syllable, and if a name has different stress, we use capital letters. This means that the English and French names ''Robert'' come out differently in Lojban: the English name is'''.robyt.''' in UK English, or '''.rab,rt.''' in some American dialects, but the French is'''.roBER.''' .


It's also handy to know that generally location and time tags come before everything else, but you need to know in what types of situations they don't. For instance, try to figure out when you want {ca ... pu ...} and when {pu ... ca ...}.
To give an idea of how all this works, here are some names of famous people in their own language and in Lojban.
 
{| class="wikitable"
Raise your awareness. But beware, you'll suddenly notice lots of mistakes! Maybe you don't want to eat of that tree.
! colspan="2" | '''English'''
|-
| ''Margaret Thatcher''
| '''.magryt.tatcys.'''


Anyway, in this fashion every possible scenario can be put into a category.
(no ''th'' in Lojban because most people around the world can't say it!)
 
|-
I'm not going to lay out all those possibilities here. I think it's good to think about this topic for yourself and to compare the different meanings the different scopes produce.
| ''Mick Jagger''
 
| '''.mik.djagys.'''
Seeing so many scope mistakes every day is slightly painful for me, because I currently have a very high awareness of the subject matter. It's easy to overlook, and I wasn't always aware of it either. Heck, only recently did I realize that my staple {tai ... ja'e ...} has inccorect scope. I'm sure there are many more constructions out there that need to be reconsidered. I'm not afraid to re-adjust my Lojban, and I recommend everyone to do the same.
|-
 
! colspan="2" | '''French'''
Lojban tends to be left-grouping, tanru and connectives for example are left-grouping. However, scope is ... "right-grouping". I'm not sure if others conceptualize it this way, but one way to parse multiple scopes is to consider each scope operator to create a sort of bubble to its right, and those bubbles are right-grouping. Abstractly:
|-
Abstractly:
| ''Napoleon Bonaparte''
<pre>⎛              ⎛                      ⎛          ⎞⎞⎞
| '''.napole,ON.bonaPART.'''
.i op1⎜  A  op2 ⎜  broda  B  op3⎜  C  D  ⎟⎜⎜
|-
⎝                  ⎝                      ⎝          ⎠⎠⎠
| ''Juliette Binoche''
 
| '''.juLIET.binOC.'''
</pre>
|-
op = phrase operator,  capital letters represent constants (i.e. unquantified sumti), the brackets should speak for themselves.
! colspan="2" | '''Chinese'''
|-
| ''Laozi''
| '''.laudz.'''
|-
| ''Mao Zedong''
| '''.maudzyDYNG.'''
|-
! colspan="2" | '''Turkish'''
|-
| ''Mustafa Kemal''
| '''.MUStafas.keMAL.'''
|-
| ''Erkin Koray''
| '''.erkin.korais.'''
|-
! colspan="2" | '''German'''
|-
|| ''Friedrich Nietzsche''
|| '''.fridrix.nitcys.'''
|-
| ''Clara Schumann''
| '''.klaras.cuman.'''
|-
! colspan="2" | '''Spanish'''
|-
| ''Isabel Allende''
| '''.izaBEL.aiendes.'''
|-
| ''Che Guevara''
| '''.tcegevaras.'''
|}


Obviously, I think scope should be given a lot more importance. It is the thing that determines the structure and word order of sentences. By acknowledging it, we can get closer to Lojban's true word order.
=== tanru<span id="2.4"></span> ===
The default grouping in Lojban is leftwards. This means that, if you have three things connected together in Lojban, the first two go together before you join in the third. For example, '''la .jasmin. je la .alis. jonai la .ranjit''' means not ''Jasmine and either Alice or Ranjeet'', but ''Either Jasmine and Alice, or Ranjeet.''
 
Does the distinction matter? Depends on your background; programmers, for example, are often driven to distraction in making sure their logical connectives work out in the right order (usually by copious use of brackets.) But there is often a real difference in meaning; the first interpretation given above describes a couple, for example, but the second doesn't.


=== Vocatives ===
melbi cmalu ractu kevna
*When it isn't your turn to speak, but you want to barge in anyway, you can say '''ta'a''' — though it probably won't make anyone any happier that you're interrupting.
*Finally, to close communication (radio's ''Over and out!''), you can use '''fe'o'''. (This is what people actually <u>should</u> be putting at the end of their e-mails; but it's not as well-known a word as '''co'o''')


=== Lambda variable in infinitives for logicians and mathematicians ===
=== Tatoeba-like. Examples of diff.words ===
{{notci|Note: To assume that Lojban works like English in general is a sin Lojbanists are ever on the alert for. It is enough of a community obsession that the Lojban word for it —'''malglico''' - ''damned English'' — routinely turns up in the English of Lojbanists, even when they're not talking about Lojban. In this instance, it is malglico to assume that ninmu refers to an adult.}}


=== mekso saske ===
{{notci|This is ''very'' Lojbanic — the English word ''good'' on its own is so vague as to be almost meaningless. It is also slightly malglico to put a person in the x1 place, which is normally filled by an object, state or event. For ‘morally good’ you would usually use vrude ‘virtuous’.}}
Likewise, you can attach a sumtcita: le nu darxi kei be gau do: “The event of hitting, which is caused by you”. Note that the presence or absence of kei makes it parse differently: With the fa'orma'o present, be attaches to nu, without the fa'orma'o, it attaches to darxi. So it decides what is being emphasised: Is the hitting, or the event of hitting caused by you? In this specific case, though, that's just about the same thing.


famyma'o Warning!: One could have description nouns inside description nouns, saying '''le le se cinjikca be mi ku gerku, = le gerku pe le se cinjikca be mi =“the dog of the man I'm flirting with”, but that's not very easy to read (or to understand when spoken), and is often being avoided.'''
:'''tu cipni la .serinus.serinus.kanarias.'''
:''<nowiki>That is a bird of species 'Serinus serinus canaria'</nowiki>''.


One need also to learn tu'a, since it will make a lot of sentences much easier. It takes a sumti and converts it to another sumti - an elliptical abstraction which has something to do with the first sumti. For example, I could saymi djica lo nu mi citka lo plise, or I could let it be up to context what abstraction about the apple I desire and just say mi djica tu'a lo plise. One always has to guess what abstraction the speaker means by tu'a SUMTI, so it should only be used when context makes it easy to guess. Another example: gasnu “x1 does/brings about x2 (volition not implied)” za'a do gasnu tu'a lo skami - “I see that you make the computer do something”. Officially, tu'a SUMTI is equivalent to le su'u SUMTI co'e. Vague, but useful. There are situations where you cannot use tu'a, even though it would seem suitable. These situations are when I don't want the resulting sumti to be an abstraction, but a concrete sumti. In this case, one can use zo'e pe. tu'a convert sumti to vague abstraction involving the sumti. Equivalent to le su'u SUMTI co'e kei ku
:'''mi pritu lo nixli lo nanla.'''
:''I'm to the right of the girl, facing the boy.''


Finally, one kind of sumti can be turned into another by the words of the class LAhE. lu'a -Convert individual(s)/mass/sequence/set to individuals. lu'i - Convert individual(s)/mass/sequence/set to a set. lu'o - Convert individual(s)/mass/sequence/set to mass. vu'i - Convert individual(s)/mass/sequence/set to sequence; the order is not stated. The use of these words is straight-forwardly: Placing them before a sumti of a certain type makes a new sumti of a new type. Notice, though, that as a fourth kind of sumti, a sequence has been introduced. This is not used very often (it doesn't have its own gadri, for instance), but just included here for completion. The last two members of LAhE do not convert between types of sumti, but allows you to speak of a a sumti by only mentioning something which references to it: If one sumti A refers to a sumti B, for instance because sumti A is a title of a book, or a name, or a sentence (which always refer to something, at least a bridi), ‘'la'e SUMTI A refers to sumti B. For instance, ‘'mi nelci la'e di'ufor “I like what you just said” (not mi nelci di'u which just means "I like your previous sentence") or ‘'la'e le cmalu noltru for the book “The Little Prince”, and not some little prince himself. The cmavo ‘'lu'e does the exact reverse – ‘'lu'e SUMTI'' refers to an object which refers to the sumti. la'e - “the thing referred to by” - extracts a sumti A from a sumti B which refers to A. lu'e - “the thing referring to” - extracts a sumti B from a sumti A, when B refers to A.''
:'''la .pybysys. cu tivni la .niksyn.in.tcainas. la .kycy,etys. la .telis.'''
:''PBS (the American Public Broadcasting Service) televises 'Nixon in China' (an opera) through KCET (the Los Angeles PBS affiliate) to Telly (a pet name for a particular television) (!)''.


----
:'''do klama la .uacintyn. la .losandjeles. la .cikagos. la .amtrak.'''
In order to clarify that all tenses are relative to the speaker's current position, the word'''nau''' can be used at any time. Another word, '''ki''' marks a tense which is then considered the new standard. That will be taught way later. '''nau''' updates temporal and spacial frame of reference to the speaker's current 'here and now'.
:''Julia travels to Washington from Los Angeles via Chicago on Amtrak (the American inter-city train network)''.
----
In the Lojban tense system, all tenses are prepositions, which we have conveniently just made ourselves familiar with. Okay okay, technically, tenses are slightly different from other prepositions, but this difference is almost insignificant, and won't be explained until later. In most aspects they are like all other prepositions; they are terminated by '''ku''', making it explicit that selma'o (the group) PU is terminated by '''ku'''.


*Every time '''nu''' starts an abstraction — a phrase nested inside another phrase —'''kei''' ends it.
How many sumti can a selbri describe? The number depends on the place structure of the word we use for the selbri. (There are ways of tagging on extra sumti, which we'll cover in later lessons). See jutsi!
*Every time an article like '''lo''' or '''loi''' starts a sumti, '''ku''' ends it.
*Every time a string of numbers starts, '''boi''' ends it.
*Every time a series of sumti follows a selbri, '''vau''' ends it.
*The <u>binding</u> can be terminated with be'o - one instance of be'o for each selbri which has sumti bound by be. To list them: be binds sumti or sumtcita to selbri bei binds a second, third, fourth (etc.) sumti or sumtcita to a selbri be'o ends binding to selbri
*li'u is one of the few terminators that can almost never be missed out, since that would make everything else that follows part of the quotation.
*The elidable terminator for '''sei''' is “se'u” (of selma'o SEhU); it is rarely needed, except to separate a selbri within the '''sei''' comment from an immediately following selbri (or component) outside the comment.


For example, if we see '''cu''' in a sentence, we know that what is coming up is a selbri<nowiki>; so the </nowiki>sumti before it must now be over. So we can drop the'''ku'''. (In fact, that's why '''cu''' exists in the first place: the beginning of a verb is a much more important structural break in natural languages than the end of a noun.) If a new sentence is beginning — as signalled by perhaps the most distinctively Lojbanic word, the ‘audible punctuation’ '''.i''' — then (have we covered .i yet?) there can be no more sumti from the old sentence; so we drop the '''vau'''. In fact, it is only in situations of potential ambiguity, like the sentence we've been looking at, that you'll get terminators appearing in normal Lojban usage at all.  
=== Space ===
Remember the word ti? This is part of a series ti, ta, tu, meaning roughly ‘this’, ‘that’ and ‘that over there.’ If we're talking about places rather than things, we say vi, va, vu, meaning roughly ‘here’, ‘there’ and ‘yonder’ or ‘way over there’. Again, this is determined by the thing you're talking about. If you're telling a doctor where you feel pain, ti might be the end of your toe, while if you're talking about astronomy, ti could be the solar system. We can therefore say


{{notci|Note: Remember those pesky possessive constructions from [[#3|Lesson 3]], when you couldn't flip '''lo tamne pe lo ninmu klama''' the other way around, because it was ambiguous? All you need is '''ku''' to resolve that ambiguity: '''lo lo ninmu klama ku tamne''' means 'the woman traveller's cousin', and lo lo ninmu ku klama tamne means ‘the woman's traveller cousin.
There is a whole class of cmavo that work like ri'u, and they are called FAhA-type cmavo, so named after a (somewhat non-representative) member of their class, fa'a (in the direction of). These include to'o (away from), zo'i (to the same side as), zu'a (to the left of), ne'a (next to), ne'i (within) and so on. (Again, all the space cmavo are explained in Chapter 10 of ''The Complete Lojban Language'').


Still, most Lojbanists think the flip-around is not worth the hassle of inserting that bothersome '''ku''', so you rarely see it used when the ‘possessor’ sumti is not a one-word sumti.}}(ping!)
{{notci|Note: FAhA cmavo indicate direction, but not ''motion toward'' that direction. There is a separate cmavo for that; see [[#7|Lesson 7]].}}


=== cmene ===
Getting back to daily speech, these time and space cmavo are very useful for questions. ca ma is ‘simultaneous with what?', or in other words, ‘when?' (a simpler alternative to ti'u or di'e). Similarly, vi ma means ‘at the location of what?', or ‘where?'
The final point is stress. As we've seen, Lojban words are stressed on the penultimate syllable, and if a name has different stress, we use capital letters. This means that the English and French names ''Robert'' come out differently in Lojban: the English name is'''.robyt.''' in UK English, or '''.rab,rt.''' in some American dialects, but the French is'''.roBER.''' .


To give an idea of how all this works, here are some names of famous people in their own language and in Lojban.
=== Restrictive and non-restrictive relative clauses ===
{| class="wikitable"
Combine the following pairs of sentences into single sentences. In each case, make the second sentence a relative clause modifying the underlined noun in the first sentence. The highlighted noun in the second sentence is the same as that in the first, and will turn into '''ke'a'''. You may omit '''ke'a''' if possible.
! colspan="2" | '''English'''
Example:
|-
| ''Margaret Thatcher''
| '''.magryt.tatcys.'''


(no ''th'' in Lojban because most people around the world can't say it!)
:'''.i mi viska <u>lo botpi</u> .i <u>lo botpi</u> cu culno'''
|-
:'''.i mi viska lo botpi poi culno'''
| ''Mick Jagger''
Watch out for '''vau''' which you may have to insert.
| '''.mik.djagys.'''
{|class='wikitable'
|-
|-  
! colspan="2" | '''French'''
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo nanmu cu citka loi cidjrkari'''
|-
|
| ''Napoleon Bonaparte''
|-
| '''.napole,ON.bonaPART.'''
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo cifnu cu kakne loka citka'''
|-
|
| ''Juliette Binoche''
|-
| '''.juLIET.binOC.'''
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i mi pu viska lo ninmu vi lo barja'''
|-
|
! colspan="2" | '''Chinese'''
|-  
|-
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lonu mi viska lo ninmu cu nandu'''
| ''Laozi''
|
| '''.laudz.'''
|-
|-
|'''.i mi viska va lo barja lo ninmu .i mi klama lo barja lo briju'''
| ''Mao Zedong''
|
| '''.maudzyDYNG.'''
|-
|-
|'''.i ca lonu mi klama lo barja lo briju kei mi penmi lo nanmu .i lo barja cu snanu lo briju'''
! colspan="2" | '''Turkish'''
|
|-
|-  
| ''Mustafa Kemal''
|'''.i mi viska lo kansa be lo ninmu .i lo ninmu cu dunda lo cifnu lo kansa be lo ninmu'''
| '''.MUStafas.keMAL.'''
|
|-
|-
| ''Erkin Koray''
|'''.i mi kakne lo ka citka lo cidjrkari .i lenu citka loi cidjrkari cu nandu'''
| '''.erkin.korais.'''
|
|-
|-
! colspan="2" | '''German'''
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo nanmu cu citka loi cidjrkari'''
|-
|
|| ''Friedrich Nietzsche''
|-  
|| '''.fridrix.nitcys.'''
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo cifnu cu kakne loka citka'''
|-
|
| ''Clara Schumann''
|-  
| '''.klaras.cuman.'''
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i mi pu viska lo ninmu vi lo barja'''
|-
|
! colspan="2" | '''Spanish'''
|-
|-
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lonu mi viska lo ninmu cu nandu'''
| ''Isabel Allende''
|
| '''.izaBEL.aiendes.'''
|-
|-
|'''.i mi viska va lo barja lo ninmu .i mi klama lo barja lo briju'''
| ''Che Guevara''
|
| '''.tcegevaras.'''
|-  
|'''.i ca lonu mi klama lo barja lo briju kei mi penmi lo nanmu .i lo barja cu snanu lo briju'''
|
|-
|'''.i mi viska lo kansa be lo ninmu .i lo ninmu cu dunda lo cifnu lo kansa be lo ninmu'''
|
|-
|'''.i mi kakne lo ka citka lo cidjrkari .i lenu citka loi cidjrkari cu nandu'''
|
|-
|
|}
|}


=== tanru<span id="2.4"></span> ===
=== swadesh ===
The default grouping in Lojban is leftwards. This means that, if you have three things connected together in Lojban, the first two go together before you join in the third. For example, '''la .jasmin. je la .alis. jonai la .ranjit''' means not ''Jasmine and either Alice or Ranjeet'', but ''Either Jasmine and Alice, or Ranjeet.''
Nouns
 
animal
Does the distinction matter? Depends on your background; programmers, for example, are often driven to distraction in making sure their logical connectives work out in the right order (usually by copious use of brackets.) But there is often a real difference in meaning; the first interpretation given above describes a couple, for example, but the second doesn't.
ashes
 
bark (of a tree)
melbi cmalu ractu kevna
belly
 
bird
=== Tatoeba-like. Examples of diff.words ===
blood
{{notci|Note: To assume that Lojban works like English in general is a sin Lojbanists are ever on the alert for. It is enough of a community obsession that the Lojban word for it —'''malglico''' - ''damned English'' — routinely turns up in the English of Lojbanists, even when they're not talking about Lojban. In this instance, it is malglico to assume that ninmu refers to an adult.}}
bone
 
child  (young)
{{notci|This is ''very'' Lojbanic — the English word ''good'' on its own is so vague as to be almost meaningless. It is also slightly malglico to put a person in the x1 place, which is normally filled by an object, state or event. For ‘morally good’ you would usually use vrude ‘virtuous’.}}
cloud
 
day (not night)
:'''tu cipni la .serinus.serinus.kanarias.'''
dog
:''<nowiki>That is a bird of species 'Serinus serinus canaria'</nowiki>''.
dust
 
ear
:'''mi pritu lo nixli lo nanla.'''
earth  (soil)
:''I'm to the right of the girl, facing the boy.''
egg
 
eye
:'''la .pybysys. cu tivni la .niksyn.in.tcainas. la .kycy,etys. la .telis.'''
fat  (substance)
:''PBS (the American Public Broadcasting Service) televises 'Nixon in China' (an opera) through KCET (the Los Angeles PBS affiliate) to Telly (a pet name for a particular television) (!)''.
father
 
feather (large)
:'''do klama la .uacintyn. la .losandjeles. la .cikagos. la .amtrak.'''
fire
:''Julia travels to Washington from Los Angeles via Chicago on Amtrak (the American inter-city train network)''.
fish
 
flower
How many sumti can a selbri describe? The number depends on the place structure of the word we use for the selbri. (There are ways of tagging on extra sumti, which we'll cover in later lessons). See jutsi!
fog
 
foot
=== Space ===
fruit
Remember the word ti? This is part of a series ti, ta, tu, meaning roughly ‘this’, ‘that’ and ‘that over there.’ If we're talking about places rather than things, we say vi, va, vu, meaning roughly ‘here’, ‘there’ and ‘yonder’ or ‘way over there’. Again, this is determined by the thing you're talking about. If you're telling a doctor where you feel pain, ti might be the end of your toe, while if you're talking about astronomy, ti could be the solar system. We can therefore say
grass
 
guts
There is a whole class of cmavo that work like ri'u, and they are called FAhA-type cmavo, so named after a (somewhat non-representative) member of their class, fa'a (in the direction of). These include to'o (away from), zo'i (to the same side as), zu'a (to the left of), ne'a (next to), ne'i (within) and so on. (Again, all the space cmavo are explained in Chapter 10 of ''The Complete Lojban Language'').
hair
 
hand
{{notci|Note: FAhA cmavo indicate direction, but not ''motion toward'' that direction. There is a separate cmavo for that; see [[#7|Lesson 7]].}}
head
 
heart
Getting back to daily speech, these time and space cmavo are very useful for questions. ca ma is ‘simultaneous with what?', or in other words, ‘when?' (a simpler alternative to ti'u or di'e). Similarly, vi ma means ‘at the location of what?', or ‘where?'
hold (inhand)
 
husband
=== Restrictive and non-restrictive relative clauses ===
ice
Combine the following pairs of sentences into single sentences. In each case, make the second sentence a relative clause modifying the underlined noun in the first sentence. The highlighted noun in the second sentence is the same as that in the first, and will turn into '''ke'a'''. You may omit '''ke'a''' if possible.
lake
Example:
leg
 
liver
:'''.i mi viska <u>lo botpi</u> .i <u>lo botpi</u> cu culno'''
long
:'''.i mi viska lo botpi poi culno'''
louse
Watch out for '''vau''' which you may have to insert.
man (male)
{|class='wikitable'
meat (flesh)
|-
moon
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo nanmu cu citka loi cidjrkari'''
mother
|
mountain
|-
mouth
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo cifnu cu kakne loka citka'''
name
|
narrow
|-
neck
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i mi pu viska lo ninmu vi lo barja'''
night
|
nose
|-
person
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lonu mi viska lo ninmu cu nandu'''
river
|
road
|-
root
|'''.i mi viska va lo barja lo ninmu .i mi klama lo barja lo briju'''
rope
|
salt
|-
sand
|'''.i ca lonu mi klama lo barja lo briju kei mi penmi lo nanmu .i lo barja cu snanu lo briju'''
sea (ocean)
|
skin (ofperson)
|-
sky
|'''.i mi viska lo kansa be lo ninmu .i lo ninmu cu dunda lo cifnu lo kansa be lo ninmu'''
smoke
|
snake
|-
snow
|'''.i mi kakne lo ka citka lo cidjrkari .i lenu citka loi cidjrkari cu nandu'''
star
|
stick (of wood)
|-
stone
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo nanmu cu citka loi cidjrkari'''
sun
|
tail
|-
tongue
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lo cifnu cu kakne loka citka'''
tooth (front)
|
tree
|-
water
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i mi pu viska lo ninmu vi lo barja'''
wife
|
wind (breeze)
|-
wing
|'''.i lo ninmu cu dunda lo cifnu lo nanmu .i lonu mi viska lo ninmu cu nandu'''
woman
|
woods
|-
worm
|'''.i mi viska va lo barja lo ninmu .i mi klama lo barja lo briju'''
year
|
 
|-  
Verbs
|'''.i ca lonu mi klama lo barja lo briju kei mi penmi lo nanmu .i lo barja cu snanu lo briju'''
to  blow (wind)
|
to  breathe
|-
to  burn  (intransitive)
|'''.i mi viska lo kansa be lo ninmu .i lo ninmu cu dunda lo cifnu lo kansa be lo ninmu'''
to  come
|
to  count
|-
to  cut
|'''.i mi kakne lo ka citka lo cidjrkari .i lenu citka loi cidjrkari cu nandu'''
to  die
|
to  dig
|-
to  drink
|
to  eat
|}
to  fall  (drop)far
 
to fear
=== swadesh ===
to bite
Nouns
to fight
animal
to float
ashes
to flow
bark (of a tree)
to fly
belly
to freeze
bird
to give
blood
to hear
bone
to hit
child  (young)
to hunt (game)
cloud
to kill
day (not night)
to know (facts)
dog
to laugh
dust
to lie (onside)
ear
to live
earth  (soil)
to play
egg
to pull
eye
to push
fat  (substance)
to rain
father
to rub
feather (large)
to say
fire
to scratch (itch)
fish
to see
flower
to seed
fog
to sew
foot
to sing
fruit
to sit
grass
to sleep
guts
to smell (perceive odor)
hair
to spit
hand
to split
head
to squeeze
heart
to stab (or stick)
hold (inhand)
to stand
husband
to suck
ice
to swell
lake
to swim
leg
to think
liver
to throw
long
to tie
louse
to turn (veer)
man (male)
to vomit
meat (flesh)
to walk
moon
to wash
mother
to wipe
mountain
 
mouth
Adjectives
name
back
narrow
bad
neck
big
night
cold  (weather)
nose
dirty
person
dry  (substance)
river
dull  (knife)
road
few
root
good
rope
heavy
salt
here
sand
many
sea (ocean)
near
skin (ofperson)
new
sky
old
smoke
other
snake
red
snow
right (correct)
star
right (hand)
stick (of wood)
rotten (log)
stone
sharp (knife)
sun
short
tail
small
tongue
smooth
tooth (front)
straight
tree
thick
water
thin
wife
warm (weather)
wind (breeze)
wet
wing
wide
woman
Pronouns
woods
he
worm
I
year
we
 
they
Verbs
thou/you
to  blow (wind)
Conjunctions
to  breathe
and
to  burn  (intransitive)
because
to  come
if
to  count
with (accompanying)
to  cut
Prepositions
to  die
at
to  dig
in
to  drink
Numbers
to  eat
five
to  fall  (drop)far
four
to  fear
one
to bite
three
to fight
two
to float
Question
to flow
how
to fly
what
to freeze
when
to give
where
to hear
who
to hit
Particles
to hunt (game)
Not
to kill
Colours
to know (facts)
black
to laugh
white
to lie (onside)
yellow
to live
*bilgygau, gau ko'a bilga
to play
=== Basic vocab ===
to pull
Should be completely inside the course.
to push
#прибывать, zukte, санки
to rain
#:ralte, ponse, предоставлять
to rub
#fasnu
to say
#направлять
to scratch (itch)
#Sensory: viska - viskygau - viskyzgana, sance, tinju'i, tirnyzgana
to see
#sisku - facki
to seed
#изволять
to sew
#ценить
to sing
#denpa
to sit
#stali
to sleep
#наречия
to smell (perceive odor)
#более
to spit
#po'o
to split
#mutce
to squeeze
#ли
to stab (or stick)
#Союзы
to stand
#* что; чтобы
to suck
#* если, если бы
to swell
#* то; же; итак; следовательно,
to swim
#* чем; как (при сравнении)
to think
#* или; либо
to throw
#fa'a seka'a, bu'u, ne'i, sene'i, fa'a, to'o
to tie
#перед
to turn (veer)
#seba'i ? вместо
to vomit
#вокруг / приблизительно (сколько)
to walk
#to'o
to wash
#ze'a
to wipe
#до
 
#между
Adjectives
#против
back
#ka'ai, sepi'o - с
bad
#по, согласно
big
#посредством
cold  (weather)
#для ; чтобы / за/на – «срок, сумма»
dirty
#позади ; после
dry  (substance)
# pe - about
dull  (knife)
#cau
few
#над
good
#на
heavy
#'''do'e''' - не имеет определённого значения / все остальные предлоги.
here
#mi, do, ra, mi'ai, do'o, mi, do, za'u ra мы / вы / они
many
#lo drata
near
#lo nei (self)
new
#неопределённо-личное местоимение
old
#1- who, what, which
other
#2- where, where to, whose, how
red
#3- why, for how long, for how many
right (correct)
#После успеха с вопросительными запоминать группу утвердительных:
right (hand)
#этот, это, такое
rotten (log)
#Дальше запоминается всё очень легко и быстро. Но оптимальные порядок с моей точки зрения:
sharp (knife)
#da, lo se cuxna be do, ro da, no da
short
#Приставки: anti-/co'a /za'ure'u/no'e/to'e
small
#Суффиксы:
smooth
#Покупки / Коммерция, банки и торговля
straight
#judri
thick
#cuxna
thin
#заказывать товар/услугу
warm (weather)
#предлагать товар/услугу
wet
#количество
wide
#sumji
Pronouns
#счёт бухг.
he
#papri, banxe karda лист / что-то плоское на чём можно писать(Бумага) или может быть что-то записано(Банковская карта, Страница Книги или Сайта)
I
#linto - tilju
we
#vikmi - выделять из организма
they
#sarcu
thou/you
#jinsa - toljinsa, lumci / умывальня / умывальник / стиральная машина
Conjunctions
#Числа и измерения: часть, многий, число
and
#Время дня, года: год, час, момент / моментально – immediately
because
#прояснить(сделать ярче)
if
#danlu: #cinki насекомое, #ползать,
with (accompanying)
#censa, cevni святой – табу,святое место,кошерная еда
Prepositions
# течь / жидкость  / плавать
at
#дышать
in
#жить, моя жизнь
Numbers
#tsali, ruble
five
#sepli
four
#jbena - рождать
one
#pruxi
three
#sligu/ranti/gapci
two
#barda/cmalu
Question
#coi co'o ki'e je'e u'u
how
#sinma
what
#xamgu / tugni do ...
when
#извинять - je'e
where
#ckire, curmi
who
#cusku lo xajmi
Particles
#Общее, прочее и невошедшее в пред.группы
Not
#возраст / совершеннолетний
Colours
#sraji clani
black
#darxi
white
#красивый
yellow
#продолжаться
*bilgygau, gau ko'a bilga
#plana
=== Basic vocab ===
#dinju - zdani
Should be completely inside the course.
#sipna
#прибывать, zukte, санки
#pacna - убеждать, доказывать
#:ralte, ponse, предоставлять
#оканчивать
#fasnu
#mlana
#направлять
#tarmi
#Sensory: viska - viskygau - viskyzgana, sance, tinju'i, tirnyzgana
#tasmi
#sisku - facki
#vlipa - сильный
#изволять
#закрывать (отверстие)
#ценить
#сумасшедший
#denpa
#весёлый
#stali
#жестикулировать
#наречия
#слава
#более
#помогать
#po'o
#впадина
#mutce
#начинать
#ли
#понимать
#Союзы
#сохранять
#* что; чтобы
#страна
#* если, если бы
#читать
#* то; же; итак; следовательно,
#связывать
#* чем; как (при сравнении)
#язык общения
#* или; либо
#длинный
#fa'a seka'a, bu'u, ne'i, sene'i, fa'a, to'o
#играть
#перед
#светить
#seba'i ? вместо
#кушать
#вокруг / приблизительно (сколько)
#машина
#to'o
#имя
#ze'a
#новый
#до
#думать
#между
#терять / глагол утраты объекта
#против
#позволять / глагол соц. разрешения и запрещения
#ka'ai, sepi'o - с
#полный
#по, согласно
#близкий / приблизительный
#посредством
#быстрый
#для ; чтобы / за/на – «срок, сумма»
#знать точно / извещать = информировать
#позади ; после
#сухой
# pe - about
#значить
#cau
#simsa
#над
#sampu
#на
#ciska
#'''do'e''' - не имеет определённого значения / все остальные предлоги.
#разновидность (сорт, вид, тип)
#mi, do, ra, mi'ai, do'o, mi, do, za'u ra мы / вы / они
#верх / поверхность
#lo drata
#изменять (+Покупки(Обмен/Валюта)
#lo nei (self)
#годиться
#неопределённо-личное местоимение
#время
#1- who, what, which
#pinxe
#2- where, where to, whose, how
#pilno
#3- why, for how long, for how many
#одежда
#После успеха с вопросительными запоминать группу утвердительных:
#жить
#этот, это, такое
#путь
#Дальше запоминается всё очень легко и быстро. Но оптимальные порядок с моей точки зрения:
#valsi
#da, lo se cuxna be do, ro da, no da
 
#Приставки: anti-/co'a /za'ure'u/no'e/to'e
=== Free lunch coming. ===
#Суффиксы:
'Lojban for Beginners' will meet you under a new name.
#Покупки / Коммерция, банки и торговля
 
#judri
In better taste.
#cuxna
 
#заказывать товар/услугу
In less complexity.
#предлагать товар/услугу
 
#количество
Stay tuned!
#sumji
 
#счёт бухг.
=== Free lunch coming. ===
#papri, banxe karda лист / что-то плоское на чём можно писать(Бумага) или может быть что-то записано(Банковская карта, Страница Книги или Сайта)
'The Crash Course in Lojban'. A classical work under a fresh name.
#linto - tilju
 
#vikmi - выделять из организма
'Lojban for Beginners' redesigned to meet the new demands from the world community.
#sarcu
 
#jinsa - toljinsa, lumci / умывальня / умывальник / стиральная машина
A better quality, a slimmer design.
#Числа и измерения: часть, многий, число
 
#Время дня, года: год, час, момент / моментально – immediately
Lojban as it is in year 2015.
#прояснить(сделать ярче)
 
#danlu: #cinki насекомое, #ползать,
Got your freebie pdf yet? Just click it.
#censa, cevni святой – табу,святое место,кошерная еда
 
# течь / жидкость  / плавать
http://mw.lojban.org/index.php?title=L17-01
#дышать
 
#жить, моя жизнь
The original idea by Robin Turner.
#tsali, ruble
 
#sepli
Initially revised by Nick Nicholas, the first fluent speaker of Lojban on the planet.
#jbena - рождать
 
#pruxi
Reworked by La Gleki with the help of Lojbanic community throughout year 2014.
#sligu/ranti/gapci
 
#barda/cmalu
Just one click away!
#coi co'o ki'e je'e u'u
 
#sinma
http://mw.lojban.org/index.php?title=L17-01
#xamgu / tugni do ...
 
#извинять - je'e
Let the magic start here!
#ckire, curmi
 
#cusku lo xajmi
=== About the book ===
#Общее, прочее и невошедшее в пред.группы
This textbook is or will be further inspired by:
#возраст / совершеннолетний
 
#sraji clani
[http://en.wikibooks.org/wiki/Lojban Lojban textbook in Wikibooks], BASIC English, gua\spi, [http://uiki.pandunia.info/en:start Pandunia lessons], toki pona with it's [http://en.wikibooks.org/wiki/Updated_jan_Pije%27s_lessons textbook], Sesperanto, [https://docs.google.com/spreadsheet/ccc?key=0Ahngu1CNj7wddDZBRzgwMm1EWlpKUEJRcTQtUGNCMFE&usp=drive_web#gid=28 Conceptual Semantic Categories of Lojban gismu], [[:File:Viku - An Art Language.htm|Viku]] lessons, Loglan3 textbook, Glosa textbook, Simple-E from Lognet'99, [http://lingvoforum.net/index.php/topic,42110.new/boardseen.html survival lingvo], Easy Loglan  ([http://www.loglan.org/Articles/easy-loglan-introduction.html Introduction] and [http://www.loglan.org/Articles/easy-loglan-description.html Description]),[http://www.loglan.org/Paradigms/paradigm_n.html N-paradigm], [http://en.wikipedia.org/wiki/Standard_Average_European Standard Average European], [http://www.unilang.org/course.php?res=80 Tok Pisin for beginners], [http://lingweenie.org/conlang/ConlangersThesaurus.pdf Conlanger's Thesaurus], [http://mw.lojban.org/images/8/8c/Lojban_translation.jpg Introduction to Ceqli adapted for Lojban].
#darxi
 
#красивый
=== poi problems ===
#продолжаться
 
#plana
Another method is to replace '''lo''' with '''zo'e''' and attach any number of relative clauses connecting them with '''je''':
#dinju - zdani
:'''mi klama zo'e poi gusta lo kisto je poi berti lo tcadu'''
#sipna
:''I go to something that is a restaurant of Pakistani cuisine and that is north of town.''
#pacna - убеждать, доказывать
You can use '''da'''/'''de'''/'''di''' if it is not used otherwise in the sentence (remember that the grammar of '''da''' and '''zo'e''' differs):
#оканчивать
{{mu|mi klama da noi gusta lo kisto je poi berti lo tcadu|I go to something which is a restaurant of Pakistani cuisine and that is north of town.}}
#mlana
==== ke'a ====
#tarmi
We used '''ke'a''' in
#tasmi
{{mu|do pu tavla lo be mi mensi poi ke'a nelci nai la .rikis.martin. vau lo dinske|You talked to my sister who doesn't like Ricky Martin about economics.}}
#vlipa - сильный
This '''ke'a''' replaces '''lo be mi mensi''' in the relative clause.
#закрывать (отверстие)
 
#сумасшедший
So '''ke'a''' is used in order not to repeat '''lo be mi mensi''' twice. We can't even use '''ri''' here instead of '''ke'a''' since '''ri''' refers to the last finished noun and then noun '''lo be mi mensi''' is not finished yet, since <u>the relative clause is a part of the noun to which it is attached</u>.
#весёлый
 
#жестикулировать
'''ra''' would be okay but it isn't particularly precise.
#слава
 
#помогать
=== po and po'e ===
#впадина
'''pe''' and '''ne''' are used as loose association only, like saying '''lo stizu pe mi''' — ''my chair'' about a chair which you sit on. It's not really yours, but has something to do with you. There is just a relationshop between the two nouns.
#начинать
{{gl|stizu|x1 is a chair}}
#понимать
 
#сохранять
If you want to say something is specific to that noun you can use '''po''' instead of '''pe'''.
#страна
 
#читать
'''lo gerku pe la alis''' can be applied to a street dog that Alice likes to play with.
#связывать
 
#язык общения
But '''lo gerku po la alis''' can be applied only to a dog that, for example, was bought for her by her parents.
#длинный
 
#играть
In many cases people prefer to use '''pe''' which is more vague rather than '''po'''.
#светить
 
#кушать
#машина
#имя
#новый
#думать
#терять / глагол утраты объекта
#позволять / глагол соц. разрешения и запрещения
#полный
#близкий / приблизительный
#быстрый
#знать точно / извещать = информировать
#сухой
#значить
#simsa
#sampu
#ciska
#разновидность (сорт, вид, тип)
#верх / поверхность
#изменять (+Покупки(Обмен/Валюта)
#годиться
#время
#pinxe
#pilno
#одежда
#жить
#путь
#valsi


=== Free lunch coming. ===
Some other examples:
'Lojban for Beginners' will meet you under a new name.
{{mu|lo cukta po mi|My book}}
{{mu|lo cipni po la .meilis.|Mei Li's bird}}
{{mu|la .kokakolys. po do|Your Coca-Cola}}


In better taste.
Lastly, '''po'e''' is used for innate, intrinsic connections.


In less complexity.
:'''lo birka be mi''' or '''lo birka po'e mi''' = ''my arm''
:'''lo mamta be mi''' or '''lo mamta po'e mi''' = ''my mother''


Stay tuned!
:'''lo gugde po'e mi''' = ''my home country'' (a more precise analog is '''lo gugde poi mi pu jbena bu'u ke'a''' = ''country where .i was born'')
As you can see, in many cases the second place of verb makes '''po'e''' unnecessary.


=== Free lunch coming. ===
Because of all of that '''pe''' is used very frequent compared to '''po''' and '''po'e'''.
'The Crash Course in Lojban'. A classical work under a fresh name.


'Lojban for Beginners' redesigned to meet the new demands from the world community.
Resume. We have:
:'''pe''' — restrictive relative clause. "which is associated with..."
:'''ne''' — non-restrictive relative clause. "which is associated with..."
:'''po''' — restrictive possessive relative clause. "which is specific to..."
:'''po'e''' — restrictive innate relative clause. "which inherent to..."


A better quality, a slimmer design.
=== [[L17-02]] ===


Lojban as it is in year 2015.
=== ti o ta ===
*'''ti .o ta''' = ''this if and only if that (both or none)''
**'''ti na .o nai ta''' = ''this if and only if that (both or none)'' (the same meaning as '''.o''')
*'''ti .u ta''' = ''this whether or not that (this, and perhaps that)''
**'''ti .u nai ta''' = ''this whether or not that'' (the same meaning as '''.u''')
*'''ti .o nai ta''' = ''either this or that''
**'''ti na .o ta''' = ''either this or that'' (the same meaning as '''.o nai''')


Got your freebie pdf yet? Just click it.
=== more about be and relative clauses ===
How would we say ''You talked to my sister — the one who doesn't like Ricky Martin — about economics''? Let's take it by steps:
{{mu|do pu tavla lo be mi mensi lo dinske|You talked to my sister about economics.}}
{{mu|lo be mi mensi cu nelci nai la .rikis.martin.|My sister does not like Ricky Martin.}}
{{mu|do pu tavla lo be mi mensi poi ke'a nelci nai la .rikis.martin. vau lo dinske|You talked to my sister who doesn't like Ricky Martin about economics.}}
Notice that it is possible to move '''be mi''' to the left of the verb '''mensi'''. This allows to attach '''poi''' to '''lo be mi mensi'''. In '''lo mensi be mi poi...''' the word '''poi''' will attach to '''mi''': ''the sister of me who (I) ...''.


http://mw.lojban.org/index.php?title=L17-01
Also notice that we need to close the relative clause with '''vau''' so that '''lo dinske''' belongs to the main verb of our clause: '''tavla''', not the verb '''nelci''' inside the relative clause. Otherwise, '''lo dinske''' would be a noun of '''nelci''' and not '''tavla''' — which is not really what you want.


The original idea by Robin Turner.
Here's another example:
{{mu|mi klama lo gusta be lo kisto|I go to the Pakistani restaurant.}}
{{mu|lo gusta be lo kisto cu berti lo tcadu|The Pakistani restaurant is north of town.}}
Here we can't attach '''poi berti lo tcadu''' (''that is to the north of the city'') to '''lo gusta''' as '''lo kisto''' doesn't allow us to do that. We also can't move '''lo kisto''' to the left as '''lo kisto gusta''' would create a a compound verb (tanru). In this case we just repeat '''poi''' twice so that the second '''poi''' attaches to '''lo gusta'''.


Initially revised by Nick Nicholas, the first fluent speaker of Lojban on the planet.
To attach two or more relative clauses use '''je''' (''and'') to connect them, since both clauses are supposed to be true. For example,
:'''lo be mi mensi poi na'e nelci la .rikis.martin. <u>je</u> poi do viska ke'a ca lo prulamcte'''
:''My sister who doesn't like Ricky Martin <u>and</u> whom you saw last night.''


Reworked by La Gleki with the help of Lojbanic community throughout year 2014.
{{mu|do bajra du'i lo mlatu|You run like a cat.|You run equally to the cat running}}
:'''du'i''' — preposition. "equally to …" (comparing event to event)
As an event follows '''du'i''' in this example '''du'i''' is applied to the whole event of the clause.
If the preposition is based on a verb whose x2 is an event then the preposition is applied to the whole clause where it resides:


Just one click away!
'''du'i''' is based on '''dunli''' — ''to be equal to''. We can compare only objects with objects and events with events. Since an event is attached to '''du'i''' it is appliedю


http://mw.lojban.org/index.php?title=L17-01
=== selmaho ===
''Particles'' ('''cmavo''') are short words like '''lo''', '''cu''', '''je''', '''pu''', '''ca''', '''ba''', '''mi''', '''do''', '''ti'''. They can be divided by their meaning and function into groups that are called '''selma'o'''. E.g. '''pu''', '''ca''', '''ba''' belong to the group PU. This is just an illustration, there is no need in memorizing the names of selma'o. But it's convenient to memorize them by those groups.
=== klama ===
Well, here is the full definition of '''klama''':
{{gl|klama|x1 comes/goes to destination x2 from origin x3 via route x4 using means/vehicle x5}}


Let the magic start here!
So
 
{{gl|lo klama|a goer}}
=== About the book ===
{{gl|lo se klama|a destionation gone to}}
This textbook is or will be further inspired by:
{{gl|lo te klama|an origin, starting point gone from}}
{{gl|lo ve klama|a route}}
{{gl|lo xe klama|a vehicle}}


[http://en.wikibooks.org/wiki/Lojban Lojban textbook in Wikibooks], BASIC English, gua\spi, [http://uiki.pandunia.info/en:start Pandunia lessons], toki pona with it's [http://en.wikibooks.org/wiki/Updated_jan_Pije%27s_lessons textbook], Sesperanto, [https://docs.google.com/spreadsheet/ccc?key=0Ahngu1CNj7wddDZBRzgwMm1EWlpKUEJRcTQtUGNCMFE&usp=drive_web#gid=28 Conceptual Semantic Categories of Lojban gismu], [[:File:Viku - An Art Language.htm|Viku]] lessons, Loglan3 textbook, Glosa textbook, Simple-E from Lognet'99, [http://lingvoforum.net/index.php/topic,42110.new/boardseen.html survival lingvo], Easy Loglan  ([http://www.loglan.org/Articles/easy-loglan-introduction.html Introduction] and [http://www.loglan.org/Articles/easy-loglan-description.html Description]),[http://www.loglan.org/Paradigms/paradigm_n.html N-paradigm], [http://en.wikipedia.org/wiki/Standard_Average_European Standard Average European], [http://www.unilang.org/course.php?res=80 Tok Pisin for beginners], [http://lingweenie.org/conlang/ConlangersThesaurus.pdf Conlanger's Thesaurus], [http://mw.lojban.org/images/8/8c/Lojban_translation.jpg Introduction to Ceqli adapted for Lojban].
=== Time of day ===


=== na'e fadni gerna ===
<--Let's imagine, though, that the time is not ten past eleven, but ten <u>to</u> eleven. We can say '''li pa no pi'e pano''' (''10:50''), but we can also say '''li pa pa pi'e ni'u pano''', where '''ni'u''' is the Lojban minus sign (for negative numbers, not for subtraction) — what we are saying is ''11:−10''.
zoi .naljbo. la sampu .lojban. .naljbo. zo'u terzu'e xu


mu'o mi'e .[[User:iesk|iesk]]. 20:33, 4 septembero 2013 (UTC)
For ''half past eleven'' we can also use '''pi''' and say '''li pa pa pimu''' ''11.5''. I don't particularly like this method, but it is perfectly good Lojban. If we are using numbers for times, it is normal to use the 24-hour system, so 6 PM is '''li pa bi''' (''18:00'').-->


:i ie mi ctuca lo xorxo gerna jenai lo catni gerna
=== tails ===
:
:[[User:Gleki|Gleki]] ([[User talk:Gleki|talk]]) 11:08, 5 septembero 2013 (UTC)


:: .i .e'o do jai gau tcita co kajde be fi lo du'u lo lobybau be do cu drata lo lobybau be lo'e se lobybau mu'o mi'e .[[User:iesk|iesk]]. 15:30, 8 septembero 2013 (UTC)
23:32 < durka42> gerna: broda gi'e ca da brode                                 
23:32 < gerna> not grammatical: broda gi'e ca _da_ ⚠  brode                   
23:33 < zahlman> what were you trying to express?                             
23:33 < durka42> ko'a ge broda gi ca da brode                                 
23:35 < durka42> I thought the tag shouldn't interact with the connective     
                unless I used {gi'ecabo} which I didn't                       
23:38 < zahlman> gerna mi ca da broda                                         
23:38 < gerna> (0[{mi <ca da>} CU {broda VAU}])0                               
23:38 < zahlman> notice that {ca da} is not part of the selbri here           
23:38 < zahlman> gi'e connects bridi-tails, so they have to start with selbri.


vi'o i ku'i caku le cukta pu'o mulno i e'anairu'e pilno ri lo nu cilre tu'a lo jbobau [[User:Gleki|Gleki]] ([[User talk:Gleki|talk]]) 17:23, 10 septembero 2013 (UTC)
=== member vs. full set ===


=== poi problems ===
<pre>
 
10:04 < latro`a> gleki: I think in the gismu themselves, it is best if these are polymorphic
Another method is to replace '''lo''' with '''zo'e''' and attach any number of relative clauses connecting them with '''je''':
10:04 < latro`a> then you can specify later
:'''mi klama zo'e poi gusta lo kisto je poi berti lo tcadu'''
10:05 < latro`a> unfortunately this does lead to odd interaction with quantifiers, because polymorphic selbri really are polyvalent
:''I go to something that is a restaurant of Pakistani cuisine and that is north of town.''
10:05 < latro`a> so {mi po'o se bangu la lojban} makes sense for one of the manifestations of {bangu}
You can use '''da'''/'''de'''/'''di''' if it is not used otherwise in the sentence (remember that the grammar of '''da''' and '''zo'e''' differs):
10:06 < latro`a> in haskell this problem is solved by the typechecker: a polymorphic function comes with a dictionary that maps values of the variable input types to implementations
{{mu|mi klama da noi gusta lo kisto je poi berti lo tcadu|I go to something which is a restaurant of Pakistani cuisine and that is north of town.}}
10:07 < latro`a> for example, if I define "f x = 3*x+1", I get a function that carries around a dictionary that tells it to use (*) and (+) from Int if x is an Int, from Double if x is a Double, etc.
==== ke'a ====
10:08 < latro`a> it'd be interesting if we could somehow implement something like this
We used '''ke'a''' in
10:08 < latro`a> but we probably can't, which means you really shouldn't use quantifiers in polymorphic places
{{mu|do pu tavla lo be mi mensi poi ke'a nelci nai la .rikis.martin. vau lo dinske|You talked to my sister who doesn't like Ricky Martin about economics.}}
10:16 < latro`a> I suppose to a limited extent this could be done with SE, though this would get rather cumbersome if we had more distinctions than distributive/nondistributive and exhaustive/nonexhaustive
This '''ke'a''' replaces '''lo be mi mensi''' in the relative clause.
10:17 < latro`a> or NAhE
 
10:24 < gleki> latro`a: what should i use {poi se gunma} and {poi cmima}?
So '''ke'a''' is used in order not to repeat '''lo be mi mensi''' twice. We can't even use '''ri''' here instead of '''ke'a''' since '''ri''' refers to the last finished noun and then noun '''lo be mi mensi''' is not finished yet, since <u>the relative clause is a part of the noun to which it is attached</u>.
10:24 < latro`a> that's not enough, unfortunately, you have to tweak the selbri instead
 
10:25 < latro`a> because this isn't really about the type of the argument
'''ra''' would be okay but it isn't particularly precise.
10:25 < latro`a> skina with skina4=nonexhaustive and skina with skina4=exhaustive are different predicates
 
10:25 < gleki> to tweak? how?
=== po and po'e ===
10:26 < gleki> for skina i can say {mi catlu lo skina}
'''pe''' and '''ne''' are used as loose association only, like saying '''lo stizu pe mi''' — ''my chair'' about a chair which you sit on. It's not really yours, but has something to do with you. There is just a relationshop between the two nouns.
10:26 < latro`a> NAhE or SE are the syntax you want
{{gl|stizu|x1 is a chair}}
10:26 < gleki> but about the other cases?
 
10:26 < latro`a> I'm imagining a NAhE for "x1 is exhaustive"
If you want to say something is specific to that noun you can use '''po''' instead of '''pe'''.
10:26 < latro`a> then "x2 is exhaustive" is {se NAhE* se}
 
10:26 < gleki> "this is a language of French people" = ko'a bangu lo fraso
'''lo gerku pe la alis''' can be applied to a street dog that Alice likes to play with.
10:27 < gleki> "this is a language of some French people (my friends)" = ko'a bangu lo fraso
 
10:27 < gleki> how to distinguish them?
But '''lo gerku po la alis''' can be applied only to a dog that, for example, was bought for her by her parents.
10:30 < gleki> seriously i understood nothin from ur proposal
 
10:33 < latro`a> ko'a se NAhE* se bangu mi == ko'a is a language of me exclusively
In many cases people prefer to use '''pe''' which is more vague rather than '''po'''.
10:33 < latro`a> where NAhE* is a new NAhE
 
10:33 < latro`a> this is a bit stronger than the {po'o} version
 
10:44 < Ilmen> coi. What do you mean by «exhaustive»?
Some other examples:
10:48 < gleki> cipra: ko'a se na'e se bangu mi
{{mu|lo cukta po mi|My book}}
10:48 < cipra> (ko'a [CU {se <na'e (¹se bangu¹)>} {mi VAU}])
{{mu|lo cipni po la .meilis.|Mei Li's bird}}
10:53 < gleki> that would be nice ofc. to have them polymorphic but then what would be the difference between cmima and se gunma? ;)
{{mu|la .kokakolys. po do|Your Coca-Cola}}
10:59 < latro`a> cmima isn't really a thing in post-set jbo
 
10:59 < gleki> what is it then?
Lastly, '''po'e''' is used for innate, intrinsic connections.
10:59 < latro`a> cmima?
 
10:59 < gleki> yes
:'''lo birka be mi''' or '''lo birka po'e mi''' = ''my arm''
10:59 < latro`a> it's the way you refer to sets
:'''lo mamta be mi''' or '''lo mamta po'e mi''' = ''my mother''
10:59 < latro`a> which we don't use
 
11:00 < gleki> so now we dont need {cmima} gismu at all?
:'''lo gugde po'e mi''' = ''my home country'' (a more precise analog is '''lo gugde poi mi pu jbena bu'u ke'a''' = ''country where .i was born'')
11:00 < latro`a> not really no
As you can see, in many cases the second place of verb makes '''po'e''' unnecessary.
11:00 < latro`a> se gunma is mass membership
 
11:00 < gleki> So "I am a member of this committee" would be ...
Because of all of that '''pe''' is used very frequent compared to '''po''' and '''po'e'''.
11:00 < gleki> en: kamni
 
11:00 < mensi> kamni = x1 (mass) is a committee with task/purpose x2 of body x3. |>>> Board of directors/trustees/cabinet (= trukamni, gritrukamni). See also bende, kagni. |>>> noralujv
Resume. We have:
11:01 < gleki> would be what?
:'''pe''' — restrictive relative clause. "which is associated with..."
11:02 < latro`a> mi se gunma ti noi kamni
:'''ne''' — non-restrictive relative clause. "which is associated with..."
11:02 < Ilmen> So, what is an exhaustive tersu'i? Do you mean things like gunma2 which doesn't require to list all the members of the mass?
:'''po''' — restrictive possessive relative clause. "which is specific to..."
11:02 < latro`a> being literal
:'''po'e''' — restrictive innate relative clause. "which inherent to..."
11:02 < latro`a> maybe this will help
 
11:02 < gleki> latro`a: do you think i should remove {cmima} from my analog of L4B ?
=== [[L17-02]] ===
11:02 < latro`a> mi se gunma mi joi do
 
11:02 < latro`a> mi na se NAhE* se gunma mi joi do
=== ti o ta ===
11:03 < latro`a> mi jo'u do se NAhE* se gunma mi joi do
*'''ti .o ta''' = ''this if and only if that (both or none)''
11:03 < latro`a> this also gives you a way to extract the individual-blob from a mass, provided that makes sense
**'''ti na .o nai ta''' = ''this if and only if that (both or none)'' (the same meaning as '''.o''')
11:03 < gleki> why {na} in the first sentence?
*'''ti .u ta''' = ''this whether or not that (this, and perhaps that)''
11:03 < latro`a> {lo se NAhE* se gunma be ...} is unambiguously all of the individuals inside, in one lo-blob
**'''ti .u nai ta''' = ''this whether or not that'' (the same meaning as '''.u''')
11:04 < gleki> could u at least give translations?
*'''ti .o nai ta''' = ''either this or that''
11:04 < latro`a> {mi se gunma mi joi do} = I am a member of the mass {mi joi do}
**'''ti na .o ta''' = ''either this or that'' (the same meaning as '''.o nai''')
11:04 < Ilmen> What means {lo ro se gunma be ko'a}?
11:07 < latro`a> {mi na NAhE* se gunma mi joi do} = I do not satisfy "x1 (individuals) is the collection of all individual members of x2"
11:05 < latro`a> it is a slightly more vague version of the same thing
11:05 < latro`a> "all of the members of ko'a that we are talking about"
11:05 < latro`a> but we might not be talking about *all* of them
11:05 < latro`a> this NAhE* is sort of an explicit UD-twiddler
11:06 < gleki> this is one of the three questions that stop me from completing a new gimste with examples, te sumti interactions, variable types, blackjack and hookers
11:06 < latro`a> {lo ro ve skina} may not contain the entire audience of the film, because we might just be talking about a small group of friends that watch movies together
11:07 < Ilmen> To me, skina4 is the *intended* audience, not the people who actually watch skina1
11:07 < Ilmen> ie pei
11:07 < gleki> as for "intended" te sumti this is actually my second question :D
11:07 < latro`a> to me that's the exhaustive skina4, but I regularly use ve skina for "watch"
11:08 < latro`a> I do not really like kinzga
</pre>


=== independent nai is experimental ===
=== necessary and sufficient ===
{{mu|— nai xu do nelci lo gerku<br>— je'u|- You don't like dogs, do you?<br>— I don't.}}
=== more about be and relative clauses ===
How would we say ''You talked to my sister — the one who doesn't like Ricky Martin — about economics''? Let's take it by steps:
{{mu|do pu tavla lo be mi mensi lo dinske|You talked to my sister about economics.}}
{{mu|lo be mi mensi cu nelci nai la .rikis.martin.|My sister does not like Ricky Martin.}}
{{mu|do pu tavla lo be mi mensi poi ke'a nelci nai la .rikis.martin. vau lo dinske|You talked to my sister who doesn't like Ricky Martin about economics.}}
Notice that it is possible to move '''be mi''' to the left of the verb '''mensi'''. This allows to attach '''poi''' to '''lo be mi mensi'''. In '''lo mensi be mi poi...''' the word '''poi''' will attach to '''mi''': ''the sister of me who (I) ...''.


Also notice that we need to close the relative clause with '''vau''' so that '''lo dinske''' belongs to the main verb of our clause: '''tavla''', not the verb '''nelci''' inside the relative clause. Otherwise, '''lo dinske''' would be a noun of '''nelci''' and not '''tavla''' — which is not really what you want.
<pre>
 
13:38 < Ilmen> does banzunu = se sarcu?
Here's another example:
13:41 < latro`a> this is muddled by classical logic shenanigans
{{mu|mi klama lo gusta be lo kisto|I go to the Pakistani restaurant.}}
13:41 < latro`a> in classical logic, p => q (that is, ~p v q) is the same as p |- q
{{mu|lo gusta be lo kisto cu berti lo tcadu|The Pakistani restaurant is north of town.}}
13:41 < latro`a> the latter is "one can infer q from p"
Here we can't attach '''poi berti lo tcadu''' (''that is to the north of the city'') to '''lo gusta''' as '''lo kisto''' doesn't allow us to do that. We also can't move '''lo kisto''' to the left as '''lo kisto gusta''' would create a a compound verb (tanru). In this case we just repeat '''poi''' twice so that the second '''poi''' attaches to '''lo gusta'''.
13:42 < latro`a> this is the definition of |-, more or less
 
13:43 < latro`a> in general necessary/sufficient is about |-, not =>
To attach two or more relative clauses use '''je''' (''and'') to connect them, since both clauses are supposed to be true. For example,
13:43 < latro`a> and |- is different in nonclassical logics, and is also different in human discourse
:'''lo be mi mensi poi na'e nelci la .rikis.martin. <u>je</u> poi do viska ke'a ca lo prulamcte'''
13:44 < Ilmen> So in this paradigm sufficiency is janai?
:''My sister who doesn't like Ricky Martin <u>and</u> whom you saw last night.''
13:45 < latro`a> in classical logic, "p is sufficient for q" is "p => q", which is to say "p na ja q"
 
13:45 < latro`a> while "p is necessary for q" is "p <= q" which is to say "p ja nai q"
{{mu|do bajra du'i lo mlatu|You run like a cat.|You run equally to the cat running}}
13:45 < Ilmen> What about the lojban "sarcu", should it be a mere logical connective, or should it express causation?
:'''du'i''' — preposition. "equally to …" (comparing event to event)
13:46 < latro`a> in human discourse, |- is more complicated than any simple logical operator
As an event follows '''du'i''' in this example '''du'i''' is applied to the whole event of the clause.
13:46 < latro`a> because we do not say "if the sky is red then mars is blue"
If the preposition is based on a verb whose x2 is an event then the preposition is applied to the whole clause where it resides:
13:46 < latro`a> even though under the classical definitions this is true
 
13:46 < Ilmen> I'm not very familiar with |-. It's a metalogical operator?
'''du'i''' is based on '''dunli''' — ''to be equal to''. We can compare only objects with objects and events with events. Since an event is attached to '''du'i''' it is appliedю
13:46 < latro`a> |- is inference
13:47 < latro`a> xatra: tell Ilmen the name for the symbol |- in english is "turnstile"
13:47 < latro`a> bah
13:48 < latro`a> Ilmen: the name for the symbol |- in english is "turnstile"
13:48 < Ilmen> got disconnected .oi
13:48 < Ilmen> did I miss something?
13:48 < latro`a> Ilmen: the name for the symbol |- in english is "turnstile"
13:51 < Ilmen> je'e
13:51 < Ilmen> So, what {sarcu} means?
13:52 < latro`a> there's causality there, definitely
13:52 < latro`a> but it's event-causality
13:53 < Ilmen> nibli is more a sufficient condition or a necessary conditiont?
13:53 < latro`a> nibli is "x1 is logically sufficient for x2"
13:53 < Ilmen> ba'a ie
13:54 < Ilmen> {banzunu} is the sufficient version of sarcu
13:55 < Ilmen> We lack a necessary-nibli
13:55 < Ilmen> ja'o
13:59 < Ilmen> X sarcu Y = lonu X tolfau cu banzunu lonu Y tolfau  (if I'm not mistaken)
</pre>
 
=== take any apple (not more than one) ===


=== selmaho ===
14:01 < menli> {lo ro plise goi ko'a zo'u ge ro da poi me ko'a zo'u do zifre lo ka lebna da gi ko'oi do lebna pa ko'a} -- what's wrong with this?
''Particles'' ('''cmavo''') are short words like '''lo''', '''cu''', '''je''', '''pu''', '''ca''', '''ba''', '''mi''', '''do''', '''ti'''. They can be divided by their meaning and function into groups that are called '''selma'o'''. E.g. '''pu''', '''ca''', '''ba''' belong to the group PU. This is just an illustration, there is no need in memorizing the names of selma'o. But it's convenient to memorize them by those groups.
=== klama ===
Well, here is the full definition of '''klama''':
{{gl|klama|x1 comes/goes to destination x2 from origin x3 via route x4 using means/vehicle x5}}


So
{{notci|Lojban doesn't force you to use any particular philosophy, religion or political beliefs.}}
{{gl|lo klama|a goer}}
{{gl|lo se klama|a destionation gone to}}
{{gl|lo te klama|an origin, starting point gone from}}
{{gl|lo ve klama|a route}}
{{gl|lo xe klama|a vehicle}}


=== Time of day ===
=== Explain nai ===


<--Let's imagine, though, that the time is not ten past eleven, but ten <u>to</u> eleven. We can say '''li pa no pi'e pano''' (''10:50''), but we can also say '''li pa pa pi'e ni'u pano''', where '''ni'u''' is the Lojban minus sign (for negative numbers, not for subtraction) — what we are saying is ''11:−10''.
Unlike when used after verbs and pronouns '''nai''' after interjections means not just "no" but the opposite attitude.


For ''half past eleven'' we can also use '''pi''' and say '''li pa pa pimu''' ''11.5''. I don't particularly like this method, but it is perfectly good Lojban. If we are using numbers for times, it is normal to use the 24-hour system, so 6 PM is '''li pa bi''' (''18:00'').-->
But explain after BAI


=== objects dont exist ===
=== Variable types ===
12:32 <@xalbo> Some object/event places are events with possible raising, but some (what I've called Objects of Cognition) really aren't. Although those hold a lot more than just object/event.
12:34 < zahlman> "objects of cognition"?
12:38 < menli> For example, zmadu-1&2, drata-1&2, melbi-1, simsa-1&2 have no requirement on the sumti type
12:42 <@xalbo> pensi2 is the prototype OoC, for me. Similarly, melbi1. "Anything the human mind can conceive of and opine about."
12:43 <@xalbo> {zmadu} is a more complicated thing, since zmadu1 and zmadu2 do have some more restrictions (namely, that they're both compatible with the open place in the zmadu3)


[15 May 13 20:31] * gleki * objects dont exist. if one postulates that "1. objects 2. exist" then that person automatically starts searching for objects in the universe. In propoerty languages u cant even ask "Do object exist?"
=== spisa - pagbu ===
* Ilmen: piece of wood, portion of water, chunk of whatever. They are a part of a substance/mass noun/uncountable goo
* lump/cluster?
* So when you use spisa you attribute substanceness to something, and spisa1 is a chunk of it. Whereas pagbu is a part-whole relation
* My arm is a part of my body
* かたまり (n) lump; mass; bundle; clump; clod; cluster; (P)
* pagbu is partially something while spisa is part of  something?
* so some driftwood I found on the beach is a piece of wood (spisa) but if I break it in half I then have two pagbu. is that right?
* pagbu = a part of something (not necessarily delimited)


=== ze'u ===
===da===
14:07 <@xalbo> I think the point isn't that {ci da plise} is talking about three apples, it's saying that the number of apples is three. "How many apples are there?" "Three".
14:07 < mukti> So not to derail the goatlegging, but earlier I asked about whether {ro} performs existential import. The consensus seemed to be that it does not. Does this mean "The problem of 'any'" section in CLL needs an update? {ro da poi klama le zarci cu cadzu le foldi} is read to say "in fact, there are people who go to the store", and more to the point: "Lojban universal claims always imply the corresponding existential claims as wel
14:07 < mukti> l"
14:07 < selpahi> The thing is that {da} does not really make reference. If you pick out three from among the five which are all in the same place, it is pretty hard to argue that the other two are not in the domain. So I'd suggest {le} or {lo} here
14:07 < ldlework> selpahi: I think the problem is that I define the UoC with my speech
14:07 < ldlework> The state of the world, objectively, doesn't.
14:08 <@xalbo> mukti: Huh, I hadn't realized that CLL came down firmly on the side of existential import like that. Very interesting.
14:08 < selpahi> mukti: Yes, this is a known problem
14:08 < mukti> xalbo: I feel very silly asking the kinds of questions that I am today. Yeah.
14:08 < mukti> selpahi: Good to know.


'''ze'u''' and it's friends also combine with other tenses to form compound tenses. The rule for '''ze'u''' and the others are that any tenses preceding it marks an endpoint of the process (relative to the point of reference) and any tenses coming after it marks the other endpoint relative to the first:
06:43 < mukti> "I want to marry any rich woman" is different from either the
{{mu|do citka pu ze'u ba zu|You ate for a long time.|You eat beginning in the past and for a long time ending at some point far into the future of when you started}}
              specific or non-specific reading of "I want to marry a rich
 
              woman"
=== scope/hoisting ===
06:44 < gleki> is the second da poi ricfu ninmu cu jai se djica mi fai lo ka mi
 
              se speni
21:45 < agbakate> a bridi creates a new scope
06:44 < gleki> ?
21:45 < agbakate> but it inherits undeclared variables from parent scopes
06:47 < gleki> then the first is {ro ricfu ninmu ka'e jai se djica ..
 
06:48 < menli> {mi djica lo nu mi speni pa da poi ricfu gi'e ninmu} vs {pa da poi ricfu gi'e ninmu zo'u mi djica lo nu mi speni da}
=== tails ===


23:32 < durka42> gerna: broda gi'e ca da brode                                 
== Fixed issues ==
23:32 < gerna> not grammatical: broda gi'e ca _da_ ⚠  brode                   
*<u>fxd?</u>The modern interpretation of ka and of other abstractors.
23:33 < zahlman> what were you trying to express?                              
*<u>fxd?</u>That if you want a Lojban name, it can be a selbri
23:33 < durka42> ko'a ge broda gi ca da brode                                 
*<u>fxd</u> restore {le} explaining what it means? (we have {le} retaining in old l4b tables and unexplained)
23:35 < durka42> I thought the tag shouldn't interact with the connective     
*<u>fxd</u> When you cover the vowels, put .ybu in the table. It is confusing otherwise and will be more useful that way for quick reference. You can still explain the vowels as you currently do after the table.
                unless I used {gi'ecabo} which I didn't                       
*<u>fxd?</u> cmavo can be written without spaces
23:37 < zahlman> aha                                                           
*<u>fxd?</u> That vi does not mean "at location ..." (that's bu'u), but rather that it tags a spatial distance, like zi tags a temporal one.
23:38 < zahlman> gerna mi ca da broda                                         
*<u>fxd?</u> me SUMTI moi
23:38 < gerna> (0[{mi <ca da>} CU {broda VAU}])0                               
* <u>fxd?</u> Then, I would also consider renaming the chapter a bit. Many concepts mentioned do not, technically, fall under "grammar" (such as pronunciation).
23:38 < zahlman> notice that {ca da} is not part of the selbri here           
* <u>fxd?</u> Also, I maybe wouldn't call it "newbies". I'm a computational scientist myself, so I understand where the term comes from, but it is not something that has really passed into the language. I never hear somebody calling somebody else on the street a "newbie" (the "for dummies" actually uses a more recognized word).
23:38 < zahlman> gi'e connects bridi-tails, so they have to start with selbri.
* <u>fxd?</u> Other than that, it seems well structured, if a bit short at some points. Remember that concepts which probably seem natural and easy to you have never been seen before by your readers.
* <u>fxd?</u> This is a problem I noticed with wave lessons continued: Some of the things mentioned (not many, thankfully), require quite a bit of thinking and research in figuring out, because the authors assumed the readers would just understand it right away.
* <u>fxd?</u> When introducing zo «ti» and zo «tu», why not fill in the blank with zo «ta» and give them their first series? (You can also appeal to Spanish for differentiating betwixt these words, since English does not recognize such a distinction but many English speakers are familiar with Spanish, which does.)
* <u>fxd?</u> When mentioning emotions in the introduction, I presume that you are referring to attitudinals? If so, are they completely novel to human language? Also, why not throw in reference to evidentials, discursives, vocatives, and erasure? Those are all sort of bundled together in my mind because they are rather simple and stand-alone in nature, but pack a mighty punch.
* <u>fxd?</u> The introduction keeps beginning paragraphs with "Lojban". I am not sure whether it is stylistic or not. If so, leave it as it is if you want (although I would check to see that you have a prime number of "Lojban"'s at the beginning of new paragraphs- humans find such combinations aesthetically pleasing). If it was not intentional, you may consider revising the wording a bit. It can cone across as a bit clunky. (I know that this is a draft, but I thought that we might as well address it now.)
* <u>fxd</u> I would give some mention to tenses and connectives, since Lojban is rather careful and unique in how it implements them in their totalities.
* <u>fxd</u> Also, are you going to gloss translations, simply give colloquial English translations, or both? If the last, will both the gloss and the colloquial translation be italicized?


=== member vs. full set ===
== L17-05 - NO FIX issues ==
 
=== Indefinite articles ===
<pre>
In English we say "a table" and "an egg". The "a" and "an" are indefinite articles. Lojban does not use indefinite articles. It is enough to say '''lo sovda''' for egg or "an" egg; '''lo jubme''' for table or "a" table, and so on, when referring to "some unspecified" egg or table, or "any" egg or table. The definite article "the", used to specify a particular object being discussed, is replacing '''lo''' to '''le''' '''lo fetsi''' - a female, '''le fetsi''' - ''the female'' or just ''she'' (a specific female, the one we are discussing).
10:04 < latro`a> gleki: I think in the gismu themselves, it is best if these are polymorphic
=== Other vocatives ===
10:04 < latro`a> then you can specify later
'''co'o''' is the farewell word, corresponding to ''Goodbye'', ''Farewell''. Lojbanists signing off on e-mail often end with something like '''co'o mi'e .bob.''' — this is equivalent to putting your name at the end of your email in English as a signature, and translates as ''Goodbye; I'm Bob.''
10:05 < latro`a> unfortunately this does lead to odd interaction with quantifiers, because polymorphic selbri really are polyvalent
10:05 < latro`a> so {mi po'o se bangu la lojban} makes sense for one of the manifestations of {bangu}
10:06 < latro`a> in haskell this problem is solved by the typechecker: a polymorphic function comes with a dictionary that maps values of the variable input types to implementations
10:07 < latro`a> for example, if I define "f x = 3*x+1", I get a function that carries around a dictionary that tells it to use (*) and (+) from Int if x is an Int, from Double if x is a Double, etc.
10:08 < latro`a> it'd be interesting if we could somehow implement something like this
10:08 < latro`a> but we probably can't, which means you really shouldn't use quantifiers in polymorphic places
10:16 < latro`a> I suppose to a limited extent this could be done with SE, though this would get rather cumbersome if we had more distinctions than distributive/nondistributive and exhaustive/nonexhaustive
10:17 < latro`a> or NAhE
10:24 < gleki> latro`a: what should i use {poi se gunma} and {poi cmima}?
10:24 < latro`a> that's not enough, unfortunately, you have to tweak the selbri instead
10:25 < latro`a> because this isn't really about the type of the argument
10:25 < latro`a> skina with skina4=nonexhaustive and skina with skina4=exhaustive are different predicates
10:25 < gleki> to tweak? how?
10:26 < gleki> for skina i can say {mi catlu lo skina}
10:26 < latro`a> NAhE or SE are the syntax you want
10:26 < gleki> but about the other cases?
10:26 < latro`a> I'm imagining a NAhE for "x1 is exhaustive"
10:26 < latro`a> then "x2 is exhaustive" is {se NAhE* se}
10:26 < gleki> "this is a language of French people" = ko'a bangu lo fraso
10:27 < gleki> "this is a language of some French people (my friends)" = ko'a bangu lo fraso
10:27 < gleki> how to distinguish them?
10:30 < gleki> seriously i understood nothin from ur proposal
10:33 < latro`a> ko'a se NAhE* se bangu mi == ko'a is a language of me exclusively
10:33 < latro`a> where NAhE* is a new NAhE
10:33 < latro`a> this is a bit stronger than the {po'o} version
10:44 < Ilmen> coi. What do you mean by «exhaustive»?
10:48 < gleki> cipra: ko'a se na'e se bangu mi
10:48 < cipra> (ko'a [CU {se <na'e (¹se bangu¹)>} {mi VAU}])
10:53 < gleki> that would be nice ofc. to have them polymorphic but then what would be the difference between cmima and se gunma? ;)
10:59 < latro`a> cmima isn't really a thing in post-set jbo
10:59 < gleki> what is it then?
10:59 < latro`a> cmima?
10:59 < gleki> yes
10:59 < latro`a> it's the way you refer to sets
10:59 < latro`a> which we don't use
11:00 < gleki> so now we dont need {cmima} gismu at all?
11:00 < latro`a> not really no
11:00 < latro`a> se gunma is mass membership
11:00 < gleki> So "I am a member of this committee" would be ...
11:00 < gleki> en: kamni
11:00 < mensi> kamni = x1 (mass) is a committee with task/purpose x2 of body x3. |>>> Board of directors/trustees/cabinet (= trukamni, gritrukamni). See also bende, kagni. |>>> noralujv
11:01 < gleki> would be what?
11:02 < latro`a> mi se gunma ti noi kamni
11:02 < Ilmen> So, what is an exhaustive tersu'i? Do you mean things like gunma2 which doesn't require to list all the members of the mass?
11:02 < latro`a> being literal
11:02 < latro`a> maybe this will help
11:02 < gleki> latro`a: do you think i should remove {cmima} from my analog of L4B ?
11:02 < latro`a> mi se gunma mi joi do
11:02 < latro`a> mi na se NAhE* se gunma mi joi do
11:03 < latro`a> mi jo'u do se NAhE* se gunma mi joi do
11:03 < latro`a> this also gives you a way to extract the individual-blob from a mass, provided that makes sense
11:03 < gleki> why {na} in the first sentence?
11:03 < latro`a> {lo se NAhE* se gunma be ...} is unambiguously all of the individuals inside, in one lo-blob
11:04 < gleki> could u at least give translations?
11:04 < latro`a> {mi se gunma mi joi do} = I am a member of the mass {mi joi do}
11:04 < Ilmen> What means {lo ro se gunma be ko'a}?
11:07 < latro`a> {mi na NAhE* se gunma mi joi do} = I do not satisfy "x1 (individuals) is the collection of all individual members of x2"
11:05 < latro`a> it is a slightly more vague version of the same thing
11:05 < latro`a> "all of the members of ko'a that we are talking about"
11:05 < latro`a> but we might not be talking about *all* of them
11:05 < latro`a> this NAhE* is sort of an explicit UD-twiddler
11:06 < gleki> this is one of the three questions that stop me from completing a new gimste with examples, te sumti interactions, variable types, blackjack and hookers
11:06 < latro`a> {lo ro ve skina} may not contain the entire audience of the film, because we might just be talking about a small group of friends that watch movies together
11:07 < Ilmen> To me, skina4 is the *intended* audience, not the people who actually watch skina1
11:07 < Ilmen> ie pei
11:07 < gleki> as for "intended" te sumti this is actually my second question :D
11:07 < latro`a> to me that's the exhaustive skina4, but I regularly use ve skina for "watch"
11:08 < latro`a> I do not really like kinzga
</pre>
 
=== necessary and sufficient ===
 
<pre>
13:38 < Ilmen> does banzunu = se sarcu?
13:41 < latro`a> this is muddled by classical logic shenanigans
13:41 < latro`a> in classical logic, p => q (that is, ~p v q) is the same as p |- q
13:41 < latro`a> the latter is "one can infer q from p"
13:42 < latro`a> this is the definition of |-, more or less
13:43 < latro`a> in general necessary/sufficient is about |-, not =>
13:43 < latro`a> and |- is different in nonclassical logics, and is also different in human discourse
13:44 < Ilmen> So in this paradigm sufficiency is janai?
13:45 < latro`a> in classical logic, "p is sufficient for q" is "p => q", which is to say "p na ja q"
13:45 < latro`a> while "p is necessary for q" is "p <= q" which is to say "p ja nai q"
13:45 < Ilmen> What about the lojban "sarcu", should it be a mere logical connective, or should it express causation?
13:46 < latro`a> in human discourse, |- is more complicated than any simple logical operator
13:46 < latro`a> because we do not say "if the sky is red then mars is blue"
13:46 < latro`a> even though under the classical definitions this is true
13:46 < Ilmen> I'm not very familiar with |-. It's a metalogical operator?
13:46 < latro`a> |- is inference
13:47 < latro`a> xatra: tell Ilmen the name for the symbol |- in english is "turnstile"
13:47 < latro`a> bah
13:48 < latro`a> Ilmen: the name for the symbol |- in english is "turnstile"
13:48 < Ilmen> got disconnected .oi
13:48 < Ilmen> did I miss something?
13:48 < latro`a> Ilmen: the name for the symbol |- in english is "turnstile"
13:51 < Ilmen> je'e
13:51 < Ilmen> So, what {sarcu} means?
13:52 < latro`a> there's causality there, definitely
13:52 < latro`a> but it's event-causality
13:53 < Ilmen> nibli is more a sufficient condition or a necessary conditiont?
13:53 < latro`a> nibli is "x1 is logically sufficient for x2"
13:53 < Ilmen> ba'a ie
13:54 < Ilmen> {banzunu} is the sufficient version of sarcu
13:55 < Ilmen> We lack a necessary-nibli
13:55 < Ilmen> ja'o
13:59 < Ilmen> X sarcu Y = lonu X tolfau cu banzunu lonu Y tolfau  (if I'm not mistaken)
</pre>
 
=== take any apple (not more than one) ===
 
14:01 < menli> {lo ro plise goi ko'a zo'u ge ro da poi me ko'a zo'u do zifre lo ka lebna da gi ko'oi do lebna pa ko'a} -- what's wrong with this?
 
{{notci|Lojban doesn't force you to use any particular philosophy, religion or political beliefs.}}
 
=== Explain nai ===
 
Unlike when used after verbs and pronouns '''nai''' after interjections means not just "no" but the opposite attitude.
 
But explain after BAI
 
=== Variable types ===
12:32 <@xalbo> Some object/event places are events with possible raising, but some (what I've called Objects of Cognition) really aren't. Although those hold a lot more than just object/event.
12:34 < zahlman> "objects of cognition"?
12:38 < menli> For example, zmadu-1&2, drata-1&2, melbi-1, simsa-1&2 have no requirement on the sumti type
12:42 <@xalbo> pensi2 is the prototype OoC, for me. Similarly, melbi1. "Anything the human mind can conceive of and opine about."
12:43 <@xalbo> {zmadu} is a more complicated thing, since zmadu1 and zmadu2 do have some more restrictions (namely, that they're both compatible with the open place in the zmadu3)
 
=== spisa - pagbu ===
 
13:00 < acolotl> Ilmen: piece of wood, portion of water, chunk of whatever.
                They are a part of a substance/mass noun/uncountable goo
13:00 < Ilmen> lump/cluster?
13:01 < acolotl> So when you use spisa you attribute substanceness to something
13:01 < acolotl> and spisa1 is a chunk of it
13:01 < acolotl> whereas pagbu is a part-whole relation
13:01 < acolotl> My arm is a part of my body
13:01 < Ilmen> かたまり (n) lump; mass; bundle; clump; clod; cluster; (P)
13:01 < Zarutian> ilmen: pagbu is partially something while spisa is part of
                  something?
13:02 < durka42> so some driftwood I found on the beach is a piece of wood
                (spisa) but if I break it in half I then have two pagbu
13:03 < durka42> is that right?
13:03 < Ilmen> pagbu = a part of something (not necessarily delimited)
 
14:07 <@xalbo> I think the point isn't that {ci da plise} is talking about three apples, it's saying that the number of apples is three. "How many apples are there?" "Three".
14:07 < mukti> So not to derail the goatlegging, but earlier I asked about whether {ro} performs existential import. The consensus seemed to be that it does not. Does this mean "The problem of 'any'" section in CLL needs an update? {ro da poi klama le zarci cu cadzu le foldi} is read to say "in fact, there are people who go to the store", and more to the point: "Lojban universal claims always imply the corresponding existential claims as wel
14:07 < mukti> l"
14:07 < selpahi> The thing is that {da} does not really make reference. If you pick out three from among the five which are all in the same place, it is pretty hard to argue that the other two are not in the domain. So I'd suggest {le} or {lo} here
14:07 < ldlework> selpahi: I think the problem is that I define the UoC with my speech
14:07 < ldlework> The state of the world, objectively, doesn't.
14:08 <@xalbo> mukti: Huh, I hadn't realized that CLL came down firmly on the side of existential import like that. Very interesting.
14:08 < selpahi> mukti: Yes, this is a known problem
14:08 < mukti> xalbo: I feel very silly asking the kinds of questions that I am today. Yeah.
14:08 < mukti> selpahi: Good to know.
 
06:43 < mukti> "I want to marry any rich woman" is different from either the
              specific or non-specific reading of "I want to marry a rich
              woman"
06:44 < gleki> is the second da poi ricfu ninmu cu jai se djica mi fai lo ka mi
              se speni
06:44 < gleki> ?
06:47 < gleki> then the first is {ro ricfu ninmu ka'e jai se djica ..
06:48 < menli> {mi djica lo nu mi speni pa da poi ricfu gi'e ninmu} vs {pa da poi ricfu gi'e ninmu zo'u mi djica lo nu mi speni da}
 
*rai
=== Fixed issues ===
*<u>fxd?</u>The modern interpretation of ka and of other abstractors.
*<u>fxd?</u>That if you want a Lojban name, it can be a selbri
*<u>fxd</u> restore {le} explaining what it means? (we have {le} retaining in old l4b tables and unexplained)
*<u>fxd</u> When you cover the vowels, put .ybu in the table. It is confusing otherwise and will be more useful that way for quick reference. You can still explain the vowels as you currently do after the table.
*<u>fxd?</u> cmavo can be written without spaces
*<u>fxd?</u> That vi does not mean "at location ..." (that's bu'u), but rather that it tags a spatial distance, like zi tags a temporal one.
*<u>fxd?</u> me SUMTI moi
* <u>fxd?</u> Then, I would also consider renaming the chapter a bit. Many concepts mentioned do not, technically, fall under "grammar" (such as pronunciation).
* <u>fxd?</u> Also, I maybe wouldn't call it "newbies". I'm a computational scientist myself, so I understand where the term comes from, but it is not something that has really passed into the language. I never hear somebody calling somebody else on the street a "newbie" (the "for dummies" actually uses a more recognized word).
* <u>fxd?</u> Other than that, it seems well structured, if a bit short at some points. Remember that concepts which probably seem natural and easy to you have never been seen before by your readers.
* <u>fxd?</u> This is a problem I noticed with wave lessons continued: Some of the things mentioned (not many, thankfully), require quite a bit of thinking and research in figuring out, because the authors assumed the readers would just understand it right away.
* <u>fxd?</u> When introducing zo «ti» and zo «tu», why not fill in the blank with zo «ta» and give them their first series? (You can also appeal to Spanish for differentiating betwixt these words, since English does not recognize such a distinction but many English speakers are familiar with Spanish, which does.)
* <u>fxd?</u> When mentioning emotions in the introduction, I presume that you are referring to attitudinals? If so, are they completely novel to human language? Also, why not throw in reference to evidentials, discursives, vocatives, and erasure? Those are all sort of bundled together in my mind because they are rather simple and stand-alone in nature, but pack a mighty punch.
* <u>fxd?</u> The introduction keeps beginning paragraphs with "Lojban". I am not sure whether it is stylistic or not. If so, leave it as it is if you want (although I would check to see that you have a prime number of "Lojban"'s at the beginning of new paragraphs- humans find such combinations aesthetically pleasing). If it was not intentional, you may consider revising the wording a bit. It can cone across as a bit clunky. (I know that this is a draft, but I thought that we might as well address it now.)
* <u>fxd</u> I would give some mention to tenses and connectives, since Lojban is rather careful and unique in how it implements them in their totalities.
* <u>fxd</u> Also, are you going to gloss translations, simply give colloquial English translations, or both? If the last, will both the gloss and the colloquial translation be italicized?


== L17-05 - NO FIX issues ==
=== Indefinite articles ===
In English we say "a table" and "an egg". The "a" and "an" are indefinite articles. Lojban does not use indefinite articles. It is enough to say '''lo sovda''' for egg or "an" egg; '''lo jubme''' for table or "a" table, and so on, when referring to "some unspecified" egg or table, or "any" egg or table. The definite article "the", used to specify a particular object being discussed, is replacing '''lo''' to '''le''' '''lo fetsi''' - a female, '''le fetsi''' - ''the female'' or just ''she'' (a specific female, the one we are discussing).
=== Other vocatives ===
*Here are some more rare vocatives:
*Here are some more rare vocatives:
*'''co'oi''' is the greeting/parting word much like Italian ''ciao'': it corresponds to ''Hello'' / ''Bye''.
*'''co'oi''' is the greeting/parting word much like Italian ''ciao'': it corresponds to ''Hello'' / ''Bye''.
Line 2,272: Line 2,146:


Later in this course we'll see how to create new words in the form of lujvo and zi'evla and those words will have precise meanings and thus become terms. You'll actually need a pretty good knowledge of Lojban to make up lujvo on the spot, but we'll learn how to make some simple lujvo later on in this course.
Later in this course we'll see how to create new words in the form of lujvo and zi'evla and those words will have precise meanings and thus become terms. You'll actually need a pretty good knowledge of Lojban to make up lujvo on the spot, but we'll learn how to make some simple lujvo later on in this course.
=== Spelling styles. Capital letters and punctuation ===
Capital letters are optionally used in Lojban for stylistic purposes:
:'''la .Alis.''' is the same as '''la .alis.'''
Punctuation can also be used to help visually structure the text. However, punctuation is used only as a decoration. It doesn't add any meaning to the text.
The symbol '''.''' (dot) can be used as we use dot in English (i.e. as a punctuation mark) but its main purpose in Lojban is that it is a proper letter that denotes a pause.


=== aspects ===
=== aspects ===
Line 2,327: Line 2,193:


In Lojban, we also operate with an event's ''natural beginning'' and its ''natural end''. The term ''natural'' is highly subjective in this sense, and the natural end refers to the point in the process where it should end. You can say about a late train, for instance, that its process of reaching you is now extending beyond its natural end. An undercooked, but served meal, similarly, is being eaten before the natural beginning of the process.
In Lojban, we also operate with an event's ''natural beginning'' and its ''natural end''. The term ''natural'' is highly subjective in this sense, and the natural end refers to the point in the process where it should end. You can say about a late train, for instance, that its process of reaching you is now extending beyond its natural end. An undercooked, but served meal, similarly, is being eaten before the natural beginning of the process.
=== Spelling styles. Punctuation ===
Capital letters are optionally used in Lojban for stylistic purposes:
:'''la .Alis.''' is the same as '''la .alis.'''
Punctuation can be used to help visually structure the text. However, punctuation is used only as a decoration. It doesn't add any meaning to the text.
The symbol '''.''' (dot) can be used as we use dot in English (i.e. as a punctuation mark) but its main purpose in Lojban is that it is a proper letter that denotes a pause.


=== Precedence ===
=== Precedence ===
Line 2,350: Line 2,208:


There are no forethought versions of clause tail conjunctions. In practice, however, two clauses connected by '''ge''' can be clause tails just as easily as a full clause: there is no real distinction in meaning between the two.
There are no forethought versions of clause tail conjunctions. In practice, however, two clauses connected by '''ge''' can be clause tails just as easily as a full clause: there is no real distinction in meaning between the two.
=== se'i, se'inai ===
The temptation to use interjections for others' reactions is strong enough, in fact, that there are a couple of ways of getting around it. If you add the modifier '''dai''', you're saying that the emotion is someone else's, and that you are empathizing with them. If '''.a'u''' is ''That's interesting!'', '''.a'u dai''' is more like ''That must have been interesting for you!'' If you add the modifier '''se'i''', you say that you feel the emotion for yourself. If you add '''se'i nai''', then, you say that you feel it for someone else: '''.uise'inai''' is pretty much ''I'm happy for you!''


==== Situation: &quot;My garden&quot; ====
==== Situation: &quot;My garden&quot; ====
Line 2,428: Line 2,282:
{{mu|mi zgana do lo ka tirna|I observe you by hearing.}}
{{mu|mi zgana do lo ka tirna|I observe you by hearing.}}


== Syntax ==
=== Types of nouns: nouns from verbs, '''la'''-names and pronouns ===
=== Types of nouns: nouns from verbs, '''la'''-names and pronouns ===
There are three types of nouns in Lojban:
There are three types of nouns in Lojban:
Line 2,512: Line 2,365:
|}
|}


== scope ==
=== lo ===
 
*{{mun|mi na cadzu ki'u lo nu carvi|It's not that I walk when it's raining.}}
**Here, the speaker might imply that he or she likes going in the rain.
 
== lo ==


Notice that  '''ro lo''' means ''each, every'' (whereas '''ro''' means ''all'').
Notice that  '''ro lo''' means ''each, every'' (whereas '''ro''' means ''all'').

Revision as of 12:12, 23 April 2016

Prioritized ToDo list

Dictionary structure

  1. Explain tricky cases like with kau and nonce infinitives in the definitions themselves

In Lojban there are several types of nouns:

  • entity: pronouns and those that are marked with la (names), lo, le, la, loi.
  • infinitives: they start with lo ka, le ka, loi ka.
  • quotes:
  • If a place is marked as "text" then a quote like zo or lu ...li'u must go into that place.
  • in {lo nu brife cu bacru} selbri wins and bacru1 is an object
  • in {mi vecnu lo plise} selbri wins and vecnu2 is an abstraction thus the parser ends in {mi vecnu tu'a lo plise}
  • object=>any type mutce, milxe, dukse, satci, and so on. In fact any time one of the arguments is a property of another argument, there's a good chance that the type of that other argument is dictated only by the type of the ce'u in the property, This also applies to comparatives (zmadu, mleca, dunli, simsa,...) It would be easier to revise the list if you sort it by type signature rather than just alphabetically.For comparatives, the property is a property of each of two arguments, not just the first.

go'i

  • na

- xu lo pixra na srana = "Is it not true that the picture is relevant?"

- go'i = "Yes, it is not true." / - ja'a go'i = "No, it is true, it is relevant."

- xu lo pixra cu na'e srana nai = "Is the picture not-relevant?"

- go'i = "No, it is relevant." {go'i} doesn't copy attitudinals {xu} and {nai} and we are left with {lo pixra cu srana}.

More about {na}: http://mw.lojban.org/extensions/cll/15/9/

xorlo

  • {ci ractu cu broda} = {ci da poi ractu cu broda} = there are three rabbits such that each of them broda's.
  • {ci lo ractu cu broda} = among the things we refer to as {lo ractu} (with all the contextual effects of that), three of them each broda.
  • {lo ci ractu cu broda} = there are three rabbits, and among them, they broda (whether collectively, distributively, or some other way)

The "there are" in those two statements are different. The first one is "there exists", the second one is more like "we're talking about".

ci lo xarju poi vofli cu melbi
3 of the pigs that fly are beautiful
ci lo poi vofli ku'o xarju
ci lo xarju ku poi vofli

of the 3 pigs, those that fly are beautiful
re le poi ninmu ku'o mu prenu cu klama le zarci
Two women out of the five persons go to the market.

Per that reading ci lo xarju ku poi vofli cu melbi (Three fliers among the pigs are beautiful)

cirko: So lo xo'e broda poi brode is the same as lo xo'e broda je brode, right?

cirko: Apparently ci lo poi sipna ku'o gerku is synonymous to ci lo gerku ku poi sipna.

bi lo bakni *ku* poi cikna cu citka
eight things that [are among [the cows] and are awake] are eating .

So without the ku it binds to the selbri and becomes sort of an adverbial clause?

the ku poi versions asserts that the cows are awake. The ku-less version only says that the eight things among the cows are awake (but not the others ones) so basically it's ((re le mu prenu ku) poi ninmu) and (re (le mu prenu poi ninmu)). According to la tersmu lo re bakni ku poi ke'a pinxe is ambiguous as to whether ke'a refers to one or many or all of the cows.

ambigu'i

Exemplary sentences I might choose a couple from:

  • I don't drink alcohol for religious reasons. I drink it for other reasons.
  • A logician's wife is having a baby. The doctor immediately hands the newborn to the dad. The wife says, "Is it a boy or a girl?" The logician says, "Yes."
  • Me and my brother are getting married this summer.
  • The man knocked on the front door and housekeeper Sarah Lim came down the stairs in a nightgown and opened it to let the visitor in.
  • Mano met a woman with a wooden leg named Aminah.
  • The lady hit the man with an umbrella.
  • They are looking for teachers of French, German and Japanese.
  • Fred saw the plane flying over Zurich.
  • Fred saw the mountains flying over Zurich.
  • The police arrested the demonstrators because they committed violence.
  • The police arrested the demonstrators because they feared violence.
  • Mary saw the dog in the store window and wanted it.
  • Mary saw the dog in the store window and pressed her nose up against it.
  • Some children may not swim because they won't behave. May children who will go home?

Splicing in time

  1. at 2 o'clock = ti'u li re pi'e xo'e
  2. at 2:00 = ti'u li re
  3. during the first two hours = ca lo pamoi be lo cacra be li re
  4. ???next hour = ca lo remoi be lo cacra
  • de'i li pa - on [January] 1
    • ba de'i li pa ki cabu = tomorrow
    • de'i li pa poi se masti ki cabu = next month
    • ti'u li pa = at 1 o'clock
    • ba ti'u li pa ki cabu = next hour
    • ti'u li pa poi se snidu ki cabu = next second

be and selbri

lo mlatu cu sipna bu'u lo tuple be mi (a sentence from nintadni)

no need for su'u or something!

  • could someone explain me the use of "be" as in "mi nelci lo bangu be mi". {be} attaches places to nouns. E.g. {bangu} = is a language. {bangu mi} = is a language of me. But when you turn {bangu} into a noun by adding {lo} you would get two nouns: {lo bangu mi} = "language, I". To still attach it to {bangu} you use {be}.
  • http://selpahi.weebly.com/archive-pre-2014/the-x2-rule Both of the rules mentioned there are wrong. The correct rule is "if there are no sumti before the selbri, then counting starts at x2 instead of x1"
  • a verb in front of anywhere other than the beginning of the sentence. (We can't just put the selbri at the very beginning of the sentence, without fa before the x1 sumti, because this would imply ‘someone/something' for the first place: the selbri would become an observative.)
  • Consider:

(1) mi dunda no da su'o de

(2) mi dunda be no da su'o de

(1) means that I don't give anything to anyone, while I'm suggesting that (2) should mean that there's someone to whom I give nothing.

In other words, (1) is:

no da su'o de zo'u mi da de broda

with broda = "dunda", whereas (2) should be:

su'o de zo'u mi de brode

with brode = "dunda be no da", where "be no da" is part of the selbri, not part of the arguments of the main bridi.

  • <@tsani> 1) moving sumti before and after the selbri does not alter the place structure, as long as there is at least one sumti in the bridi head.

08:01 <@tsani> 2) a be-clause produces a new selbri, by partially applying it to one or more sumti. 08:01 <@tsani> (the second premise is evidenced in the formal grammar)

to have

  • Which name do you have? We simply use the predicate {cmene}:

.i mi se cmene zo tsani -> "My name is 'tsani'." .i zo .djeikyb. cmene mi -> "'Jacob' is my name."

  • For legal ownership, there's {ponse}:

.i mi ponse lo re karce -> "I own two cars." .i mi ponse lo mu mlatu -> "I own five cats."

  • For pets, it may make more sense to use {kurji} instead of {ponse}.

.i mi kurji lo ci gerku -> "I take care of three dogs."

  • For obligations, we use {bilga} with a ka-abstraction or a tanru.

.i mi bilga lo ka jibykai -> "I must work." .i mi jibykai bilga -> "I must work." .i mi bilga co cidja dunda fi lo mlatu -> "I must feed the cats."

  • For necessities, we use {nitcu}. nitcu vs bilga depends on how badly the thing is needed or the nature of the need.

.i mi nitcu lo ka citka -> "I need to eat." .i mi nitcu lo ka sipna -> "I need to sleep."

  • For physical possession of things, regardless of ownership, you can use {bevri} and {jgari}:

.i mi bevri lo samyfonxa -> "I have a smartphone." (not saying that you own it; it could be someone else's) .i mi jgari lo xance be do -> "I'm holding your hand."

  • Sumti possessives can be formed with {pe}:

{lo stizu pe do} -> "your chair", literally "the chair restrictively-related-to you" i.e. the fact that you're associated somehow to the chair is information allowing us to disambiguate among some bunch of contextually-relevant chairs. {pe} can be rendered with the selbri {srana} or with {co'e}: .i lo stizu cu srana do -> "The chair is associated with you." .i lo stizu do co'e -> "The chair has an obvious/unimportant relationship with you." ==Scope

  • mi na cadzu ki'u lo nu carvi
    It's not that I walk when it's raining.
    • Here, the speaker might imply that he or she likes going in the rain.
  • a bridi creates a new scope
    • but it inherits undeclared variables from parent scopes
  • da scope.
    • {da zo'u da broda .i je da brode} = {da zo'u ge da broda gi da brode} = {suzdza lo ka poi'i ge ke'a broda gi ke'a brode}
    • 05:03 < Ilmen> {da poi'i ge ke'a broda gi ke'a brode}

13:41 < selpahi> {ze'i so'i roi plipe} vs {so'i roi ze'i plipe}, {ze'i lo mentu mi so'i roi [ri] cu plipe} vs {so'i roi ku mi plipe ze'i lo snidu}

{puku da renvi} = there were survivors.

{da puku renvi} = Some survived.

{mi co'u ponse da} ----> tolcfa fa lo nu li su'o kaidza lo ka mi ponse ke'a - i lost everything. {da co'u se ponse mi} - there is something i lost.

I wasn't sure whether you were saying you come here when you are writing, or you don't come here often because you are usually writing

{mi ru'inai jundi mu'i lo nu finti} implies the former


"i come here rarely in order to write" - "in order to write i come here rarely"


{mi na klama mu'i tu'a lo carvi} (“It is not true that [I go motivated by the rain]”), when they actually mean {mi na'e klama mu'i tu'a lo carvi} (“[I don't go], which is motivated by the rain”). 

< durka42> gleki: the classic example of going to the bank in the rain, for one 10:30 < gleki> durka42: pardon? 10:31 < lai_az> le nu mutce nandu mi cu rinka zo'e bi'unai 10:31 < lai_az> or zo'e bi'u if its new 10:31 < lai_az> or ti etc 10:31 < durka42> gleki: mi na klama lo banxa ki'u lo nu carvi 10:32 < gleki> sounds like "i dont drink alcohol for religious reasons" 10:32 < durka42> same, yes 10:33 < gleki> but how to translate it? for me the straightforward translations is a mental puzzle 10:33 < durka42> na-ki'u: religion isn't the reason I drink alcohol; ki'u-na: due to religion I don't drink 10:46 < gleki> and with the rain? 10:46 < durka42> na-ki'u: rain doesn't draw me to the bank, or something, u'i

Following Lojban usage, I observe an extremely high number of scope mistakes. It appears that people are generally unaware. Their sentences are usually still understood "thanks" to the human capability of letting context fix scope dependencies. That's why these mistakes tend to go unnoticed. Nobody corrects these mistakes, and they keep on being made. It's a problem, but it's not too late to get a new awareness of scope.One of the most common scope mistake is this:

(1) mi na terve'u lo stagi ki'u lo nu mi na bevri lo jdini "I did not buy vegetables because I didn't have money with me."

If you just read the English, your mind is likely to autocorrect the scope, causing you to miss the problem in the Lojban sentence. Though this might seem like an obvious mistake, I just saw someone make it again today.

Sentence (1) actually means: "That I bought vegetables because I didn't have money with me is false."

That's how the Lojban should be understood, but it often isn't. A practical fix is to use {.i ki'u bo}:

(2) mi na terve'u lo stagi .i ki'u bo mi na bevri lo jdini "I didn't buy vegetables. The reason for that is that I didn't have money with me."

The other method requires forethought, which might seem annoying, but you'll soon come to realize (if you haven't already) that forethought is needed in all of Lojban the moment you begin to scope-juggle.

(3) mi ki'u lo nu na bevri lo jdini cu na terve'u lo stagi "I, because [I] didn't have money with me, did not buy vegetables."

Case in point?

There are also cases of scope mistakes that I observe even among otherwise experienced Lojbanists. This is usually the case when multiple tenses are being used. Often, people just tack on an additional tense at the end of their already-tensed phrase, and the result is usually a sentence with incorrect scope, which the speaker doesn't notice.

There is a difference between {ca pu} and {pu ca}. There is a difference between {pu co'a} and {co'a pu}. You normally cannot reverse the order of phrase operators and expect to get the same result each time. While initially a difficulty, it's also one of the strong points of Lojban. It's the very thing Lojban is supposed to be good at. To neglect this aspect is to neglect Lojban.

Learning to use this feature correctly requires one to rethink. Lojban is not English. It's not even close. It just happens to sometimes be able to mirror the word order. It quickly stops doing that once the statements get more complex, at which time it drifts more towards SOV languages. It's not unnatural for a language to require constant forethoughts. Japanese and Turkish do it, just to name two. (Also note that when I say "forethought", I'm not talking about forethought connectives, although they can help. I'm talking about the general concept of thinking ahead).

Simply put, you have to learn (practically by rote, or by making the same mistake many times and correcting it until it sticks) what operators should precede what operators in what situation. Having some mental categories of different situations is very helpful and, I assume, inevitable. Well, to be sure, there are always other methods, but this one doesn't seem so painful. To give you an example, you need to simply know that a prenexed quantifier is required in this general instruction manual sentence:

(4) "Thirdly, everyone holds on to the rope."

Suppose this is a sentence that describes a hypothetical setup that the instruction manual is trying to explain to you. We can argue that the rope is not definite. It can be any rope (so {lo} fits more than {le}), but everyone is supposed to hold on to the same rope, that's the important part. The naive translation fails:

(4a) ci mai .ei ro lo prenu cu jgari lo skori

This doesn't tell us that everyone is supposed to hold on to the same rope. And we also cannot use {le} (and even if we could, it's not really the point of this example). (4a) is vague about the rope business. So what now? As I said, this type of situation needs a prenexed quantifier:

(4b) ci mai .ei pa da poi skori zo'u: ro lo prenu da jgari

This tells us explicitly that each person is supposed to hold on to the same rope.

Yes, you can also just reverse the order of the sumti as in (5), but it would be a different kind of word order (which proves my point, but messes up the example. If you want to mirror the English, (4b) is the way to go).

(5) ci mai .ei pa skori cu se jgari ro lo prenu

This one is shorter, not requiring an explicit prenex, and no explicit {da} binding. It does require one to change the order of the arguments, though, which might not always be desirable. Sometimes, placing a sumti in the x1 achieves a certain effect, e.g. topicalization and switching arguments can potentially ruin that.

It's also handy to know that generally location and time tags come before everything else, but you need to know in what types of situations they don't. For instance, try to figure out when you want {ca ... pu ...} and when {pu ... ca ...}.

Raise your awareness. But beware, you'll suddenly notice lots of mistakes! Maybe you don't want to eat of that tree.

Anyway, in this fashion every possible scenario can be put into a category.

I'm not going to lay out all those possibilities here. I think it's good to think about this topic for yourself and to compare the different meanings the different scopes produce.

Seeing so many scope mistakes every day is slightly painful for me, because I currently have a very high awareness of the subject matter. It's easy to overlook, and I wasn't always aware of it either. Heck, only recently did I realize that my staple {tai ... ja'e ...} has inccorect scope. I'm sure there are many more constructions out there that need to be reconsidered. I'm not afraid to re-adjust my Lojban, and I recommend everyone to do the same.

Lojban tends to be left-grouping, tanru and connectives for example are left-grouping. However, scope is ... "right-grouping". I'm not sure if others conceptualize it this way, but one way to parse multiple scopes is to consider each scope operator to create a sort of bubble to its right, and those bubbles are right-grouping. Abstractly: Abstractly:

⎛              ⎛                       ⎛          ⎞⎞⎞
.i op1⎜   A   op2 ⎜  broda   B  op3⎜  C   D  ⎟⎜⎜
⎝                   ⎝                      ⎝           ⎠⎠⎠

op = phrase operator, capital letters represent constants (i.e. unquantified sumti), the brackets should speak for themselves.

Obviously, I think scope should be given a lot more importance. It is the thing that determines the structure and word order of sentences. By acknowledging it, we can get closer to Lojban's true word order.

Issues: unsorted ones

    • 1. and 2. are to be solved by using prototypes. xunre is an idealistic prototype not present in real world. Or we can remember Einstein's words that all electrons have the same mass because they are simply the same electron (projected many times into the physical Universe).
  • 22:34 < gleki> 1. there are 100 women, 10% of them are pretty. He likes 10 those women

22:35 < gleki> 2. There are 100 women, 10 % of them are pretty. He like 10% of pretty women, i.e. he likes one woman. 22:36 < durka42> remai nelci dauce'i lo ro ninmu poi melbi ku noi panomei 22:37 < durka42> camxes: -f nelci dauce'i lo ro ninmu poi melbi ku noi panomei 22:37 < camxes> (nelci [dau ce'i] [lo {ro ninmu <poi melbi>} ku] [noi {pa no} mei]) 22:38 < durka42> panono da ninmu .i pano ri melbi .i nelci ri 22:38 < durka42> ja'o nelci ro lo pano lo panono ninmu ku poi melbi 22:39 < durka42> camxes: -f ja'o nelci ro lo pano lo panono ninmu ku poi melbi 22:39 < camxes> (ja'o [nelci {ro <lo (¹pa no¹) (¹lo [{pa <no no>} ninmu] ku¹) (¹poi melbi¹)>}]) 22:39 < durka42> replace {lo pano} with {lo dauce'i} if you want 22:40 < maik_> gleki, you meant to say "He likes _the_ 12% of women who are pretty" earlier

  • 12:44 < gleki> how to express "if(broda){brode}else{brodi}" without repeating broda twice?

12:45 < b_jonas> tsanire: it is the start of {.i ga nai go'i gi brodu} 12:45 < b_jonas> tsanire: so the whole thing would be {.i brodi .i ja broda .i ga nai go'i gi brodu} 12:46 < b_jonas> and I'm asking whether that means the same as {.i brodi .i ja broda .i ga nai broda gi brodu} 12:47 < b_jonas> and if it doesn't mean that, then I wonder what you could say instead of "go'i" to refer back to just "broda", where "broda" here is placeholder for a whole bridi which could be long 12:49 < latro`a> if there's some dispute about boundaries between bridi, then you could circumlocute by reifying the selbri and assigning it 12:49 < latro`a> {.i brodi .i ja me'au lo du'u broda goi ko'a .i ga nai me'au ko'a gi brodu} 12:49 < b_jonas> latro`a: yes, you could certainly assign it with cei (if I understand it right what cei means) 12:50 < b_jonas> I don't think you have to reify it for that 12:50 < latro`a> if go'i has a problem than cei does too- 12:50 < b_jonas> just assign it to brodo with cei 12:50 < latro`a> they're essentially the same idea 12:50 < b_jonas> what... oh cr 12:50 < b_jonas> you're right 12:50 < b_jonas> it's because cei is too greedy 12:50 < b_jonas> it gobbles up more than you put it next to

  • < Ilmen> it's similar to selpa'i's proposal for a cmavo allowing "proprietor fronting" (split of a NU into an object and a property)

09:15 < mukti> So the "part of" bit of your gloss of {jai cumki} has me interested. 09:15 < danr> mukti: It should be an easily accesible sumti 09:16 < mukti> Somehow x1 participates in xFAI ? 09:16 < Ilmen> "he's easy to trick" ---> ko'a jai frili loka se tcica 09:16 < gleki> quickly spltting an abstraction into an object+ its property is one of the highly needed features in lojban. 09:16 < Ilmen> in kind of nu-split is common in English

  • «lu no da zo'u: mi smàdi lo du'u da dùnda lo jdìni li'u» frìca «lu mi smàdi lo du'u no da zo'u: da dùnda lo jdìni li'u»
  • {fi'o gunma je fa lo tadni cu sruri}?

If ca'a modifies a selbri, why not inside a lo construct? 08:05 < latro`a> darsi: that would modify the wrong selbri 08:05 < darsi> oh, right. 08:05 < latro`a> we want {ko'a ca'a mruli .i ko'e ka'e se mruli} and then to somehow join these two

  • no, I don't want erasure erasure is textual. I'd like semantic one, not syntactic erasure

23:51 erasure with {sa} is ok if I misspoke, but in this case I didn't misspoke, I was just wrong about some fact 23:51 < gleki> {mi bebna i go'i je'unai i do go'i} = I am a fool. Wait, it's not true. YOU are a fool. 23:52 gleki: yes, that negates the previous statement, but then how do I tell the next statement is something of similar scope that corrects the wrong statement, not just something unrelated? 23:53 < gleki> I need an example because .uinai I cant understand you yet. 23:55 ok, how about something like "The house is big. No, that's wrong. Actually, the house is small." 23:56 now {.i go'i .uinai} could express the "No, that's wrong." part, but how do I say "Actually,"? 23:58 < gleki> lo dinju cu barda i oi go'i je'unai i lo go'i cu cmalu je'u Day changed to 27 Mar 2014 00:00 < gleki> It might look tricky because {lo go'i} refers to the first sumti in the second sentence. And {go'i} in the second sentence refers to the phrase in the first sentence.

  • a prosumti for something actually unknown would be cool
    • 13:31 < acolotl> E.g. {mi broda Y} -> {mi broda zo'e noi ke'a ka'e se no'a}
    • 13:31 < acolotl> It can be used for a restricted "everything"
    • 13:32 < acolotl> {mi djuno ro da} could be understood as too extreme
    • 13:32 < acolotl> but {mi djuno ro Y} is "I know everything that is knowable"
    • 13:33 < acolotl> And {ro Y prami do} means "Everyone that can love you does"
    • 13:33 < acolotl> Y is just everything that could fill this place
    • 13:33 < xalbo> I think it's just more useful to assume that {da} is contextually bound, because otherwise it's completely useless.
  • There are two types of conjunctions: logical and non-logical. Logical conjunctions say something about whether and in what circumstances the two things connected are true; an example is je. Non-logical conjunctions do not deal with separate truth values, but group things together to form different kinds of units.
    • Lojban distinguishes between the logical component of conjunctions, and their attitudinal content. For example, most languages have different words for and and but. Logically, they both mean the same thing. In terms of attitude, however, they are different: but contains a connotation of contrast or unexpectedness, which and does not. So Lojban translates but in two parts: je ku'i = and however. This follows the Lojban principle of keeping content and attitude separate (e.g. .ui la .jasmin. cu klama ti has a content element — the information that Jasmine is coming here, and an attitude element — happiness).
    • Lojban also handles some of the functions of English conjunctions in other ways — as we saw, because and so are translated with prepositions, not conjunctions.
  • I understand now that that is how it works, but if "ro" and "pa" combine with other members of PA differently, then how can they be considered members of the same selma'o?
    • They behave the same way grammatically, but their interpretation, which is a semantic feature not governed directly by the grammar, depends on the exact words used. The CLL describes these interpretations:
    • Another possibility is that of combining definite and indefinite numbers into a single number. This usage implies that the two kinds of numbers have the same value in the given context:

8.18) mi viska le rore gerku

      I saw the all-of/two dogs.
      I saw both dogs.

8.19) mi speni so'ici prenu

      I am-married-to many/three persons.
      I am married to three persons (which is “many” in the circumstances).

(Chapter 18, section 8, at the bottom.)

Some lojbanists extend these interpretations in many different directions, but those usages are (sadly) nonstandard. The general rule that we can discern from this CLL excerpt is that when an exact number and an inexact number are sequenced, then the result is as if these numbers were logically connected with {.e}.

  • Explain lo nu broda pu brode - and use {lo nu broda cu pu brode}. so {lo} and {sei} require the selbri to be the last word. {nu} allows anything: prepositions, nouns, at the end
    • lo nu tcidu ca nandu - The current reading complication
    • lo nu tcidu cu ca nandu - The reading is now difficult
    • lo nu tcidu ca cu nandu - The current reading is difficult.
  • xalbo: In your section "the simplest sentences", it might be beneficial to make one of your examples an "adjective" (colors are good), to show that they're verbs too.
    • gleki: wait, adjectives are NOI
      • [20:26] xalbo: I mean using {blanu} in addition to {mlatu} and {pinxe}
        • [20:26] gleki: lo blabi mlatu cu pinxe lo ladru?
          • [20:27] xalbo: I'm saying include {blanu}, which is simply translated as "blue", but is really also a verb "to be blue".

banzu

Standard:

  • banzu = x1 (object) suffices/is enough/sufficient for purpose x2 under conditions x3.
  • sarcu = x1 (abstract) is necessary/required for continuing state/process x2 under conditions x3.
  • dukse = x1 is an excess of/too much of x2 by standard x3.

Abandoned proposal:

  • banzu = x1 (nu) suffices/is enough/sufficient for purpose x2
  • sarcu = x1 (nu) is necessary/required for continuing state/process x2
  • dukse = x1 (nu) is an excess of/too much for continuing state/process x2
  • Math:
and sarcu? how would you say x1 is necessary and sufficient?
10:43 < latro`a_> sarcu gi'e se jalge
10:46 < latro`a_> or in a more material implication sense, jalge gi'e se jalge
10:46 < latro`a_> which is basically the nu version of {jo}
10:46 < latro`a_> but then, I haven't seen "necessary and sufficient" in an informal context in a long, long time
10:47 < latro`a_> whereas in a formal context {jo} is sufficient
    • New:
      • sufsi = x1 is enough in x2 (ka/ni) for purpose x3 (nu) to take place
      • dukse = x1 is too much in x2 (ka/ni) for purpose x3 (nu) to take place

the ka form is the same as the ni form after substitution, but with dukse2/banzu2 being zo'e

10:54 < latro`a_> for example, {mi dukse lo ka ma kau zdani ce'u} = {lo zdani be mi cu dukse zo'e}
10:54 < latro`a_> but with modified emphasis
      • barda and friends do the same

Super-Advanced: Negation

Lojban presents several words for expressing different shades of negation.

na

  1. na put immediately before the main verb negates the whole clause:

    lo speni be mi cu na ninmu
    It is not true that my spouse is a woman.

    speni = x1 is married to x2
    It states nothing about what my wife is, or if I even have a wife. It only states that I do not have a wife who is also a woman.
    Similarly, if someone asks you xu do pu viska la .ian. (Did you see Yan?) you can reply na go'i (No.) even you have never heard of any Yan in your life. And you won't be really lying.

na ku

  1. na ku in the beginning of a clause negates the whole clause:

    na ku lo speni be mi cu ninmu and lo speni be mi cu na ninmu mean the same.
    It is not true that my spouse is a woman.

  2. na ku in other positions negates everything to the right of itself within the current clause.
    If I say mi na sutra tavla bau lo glibau se ja'e lo nu mi dotco, I end up negating too much, and it is not clear that I wanted to only negate that I speak fast. The sentence could suggest that I in fact speak fast because of some other reason, for instance that I speak fast in French because I'm German. To express the sentence more precisely, I need to only negate that I speak fast, and not the other things. To only negate part of a clause, na ku can be moved around the clause and placed anywhere a noun can go. It then negates any noun, main verb and preposition placed after it.
  3. na ku at the rightmost end of the clause is the same as na before the the main verb of this clause.
    Often we need our negation to be a little less powerful. In particular, it is useful to be able to say, not that the whole clause is false, but only the main verb. This means that there is some relationship between the nouns — but this main verb isn't it.

Moving na ku from the left end of the sentence and rightwards effects any quantifiers in a certain way, as can be seen by this example:

na ku ro lo remna cu verba
It's not true that: All humans are children.
su'o lo remna na ku cu verba
For at least one human it's not true that: it's a child.

See that the na ku is placed before cu, since a noun can go only before, not after the cu. Had I only used na, it would have to go after cu — but that would have negated the entire clause, meaning "It's not true that: At least one human is a child".

When the na ku is moved rightwards, any quantifier is inverted — that is: ro is turned into su'o. This is, of course, only if the meaning of the clause has to be preserved. This means that when the na ku is placed at the end of the clause, only the main verb is negated but all the nouns and prepositions are preserved, as can be seen by these three identical clauses:

na ku ro lo verba cu tadni bu'u da poi ckule
It's not true that all children are students in a school.
ckule = x1 is a school at area x2
su'o lo verba cu tadni na ku bu'u da poi ckule
Some children study in not a single school.
su'o lo verba cu tadni bu'u ro lo ckule na ku
Some children are for all schools don't study in them.

It is possible to use clause negation in all clauses, even the implicit clause of descriptive nouns. lo na prenu can refer to anything non-human, whether it be a sphinx, a baseball or the property of appropriateness.

Note: Like cu, na can indicate that a main verb is coming up, so we can just say na instead of cu na. So you can say
lo ninmu na nelci la .ian. = A woman likes Yan
without adding cu before na nelci.


Interjections nai, ru'e, sai, cu'i and na'i

  1. nai being an interjection attaches to the right of a construct and negates it.

    tu bajra mlatu nai
    That is a running non-cat.

    tu bajra nai mlatu
    That is a non-running cat.

    As you can see nai is applied only to the last verb of a compound verb.
  2. When attempting to answer: “Is the king of the USA fat?”, all of these negations fail. While it's technically correct to negate it with na, since it makes no assumptions of that is true, it's mildly misleading since it could lead the listener to believe there is a king of the USA. For these scenarios, there is a metalinguistic negator, na'i.
    na'i — metalinguistic negator. Something is wrong with assigning a truth value to the clause.
    Because na'i has the grammar of interjections.

Scalar negation

na'e attaches to the left of a construct and negates it implying the existence of a scale:

tu na'e bajra mlatu
That is a other-than-running cat.
tu bajra na'e mlatu
That is a running other-than-cat.
(making us wonder what a non-cat might be)

My spouse is not a woman (meaning that he is a male).
lo speni be mi cu na'e ninmu or lo speni be mi cu to'e ninmu.

Using scalar negation here implies that he exists, which na did not.

Another unexpected effect compared to English not:

mi na nelci ro lo gerku means It is not true that I like each dog.

However,

mi na'e nelci ro lo gerku means I don't like any dogs.

While the mechanism of na ku resembles negation in English, it can be difficult to keep track of exactly what is negated and how that affects the clause. For that reason, the construct na ku is rarely seen anywhere other than the beginning of a clause. In most cases where more specific negation is needed people resort to a scalar negation using na'e. With na'e only a certain construcut is affected whereas with na ku everything to the right of it is negated. This is the reason na ku is used rather seldom and mostly at the beginning of the sentence. In

mi na'e nelci ro gerku
I other-than-like all dogs.

there is something that can be said about me and all dogs, but it's not that I like them.

I might hate them:

mi to'e nelci ro gerku
I anti-like (= dislike) all dogs.

to'e turns a main verb into its opposite: to'e nelci is pretty much the same thing as xebnito hate.

Or I just might be neutral to them. I might write poems about them, or prescribe medicine for them, imitate them in polite company, or just might be indifferent to them. Then we'll use no'e:

mi no'e nelci ro gerku
I am neutral-as-to-liking all dogs.

But as for liking them, I don't.

Thus na'e is not a negator in the same sense as na. It doesn't state that a clause is false, but makes a positive statement that a clause is true – the same clause, but with a different main verb. This distinction is mostly academic, though. If, for example, I state mi na'e se nelciI am non-liked, I actually state that some main verb applies to me, which is also on a relevant scale with the main verb nelci. Most of the time, we assume a scale where the positions are mutually exclusive like "like — dislike — hate", so mi na'e se nelci implies mi na se nelci. Therefore, the words no'e and to'e should only be used when the main verb is placed on some obvious scale: lo speni be mi cu to'e melbiMy spouse is ugly makes sense, since we immediately know what the opposite of beautiful is, while mi klama lo to'e zdani be miI go to my opposite thing of home, while grammatical, leaves the listener guessing what the speaker's opposite of home is and should be avoided.

My spouse is not really a woman.
lo speni be mi cu no'e ninmu

The scale here is presumed to be from woman to man.

I don't speak fast in English because I'm German.
mi na'e sutra tavla bau lo glibau se ja'e lo nu mi dotco

Chat

05:06 < nargavela> BTW I should ask on the mriste how to express the PEG '/' operator; it is a must for describing Lojban syntax in Lojban
05:12 < nargavela> na broda = broda be naku
05:13 < markuja> .i lu na broda su'o da li'u na dunli lu broda be naku su'o da li'u

Negation and neutral meaning

nai is an interjection that makes a part of sentence negative in meaning. nai means no or not.

  • interjection in Lojban modifies the construct to the left of it.
mi nelci nai do
I don't like you.
I like-not you (literally). [literally]
mi nai nelci do
Not I like you.
(may be someone else likes you) [literally]
So when put after certain part of the phrase like pronoun or a verb it modifies that verb.
  • if put in the beginning of a phrase interjection modifies the whole phrase:
nai mi nelci do
It's not true that I like you.
So in the beginning of the phrase nai negates the whole phrase.
  • we can put an interjection after different parts of the same phrase shifting the meaning.

The interjection cu'i makes a part of sentence middle in its meaning.

mi nelci cu'i do
As for whether I love or hate you, I'm indifferent to you. I neither like nor hate you.

Double negation

Sometimes it's necessary to use double negation. Let's show how this works using examples from English and Chinese.

The structure 非...不可 (fēi … bùkě) is one of the most commonly used in Mandarin Chinese. It means "must"/"absolutely must"/"need to." 非 means "not"/"no" and 不可 means "not possible". It's literally translated as "not not possible."

mi na cfifa'i ra na fau da
我非批评她不可。
wǒ fēi pīpíng tā bùkě
I can't not to criticize her.
I absolutely must criticize her.
mi'ai na tadni na fau da
我们非学习不可。
wǒmen fēi xuéxí bùkě
We must study.
mi'ai tadni nai fau no da
We can't not study.

fau and if

There is some controversy about against using jo and ja nai for expressing if and only if. First you shouldn't use ja nai (if) when jo (only if) is meant. And in natural languages if is strongly tied up with notions of causality, precondition, or deduction — none of which is particularly emphasized by ja nai and jo as they are strictly logical conjunctions. For example, ja nai will give a poor rendering of It's not true that, if I'm rich, I'm happy — which is decidedly not the same thing as It's not true that I'm either not rich or happy! For that reason, you will see many Lojbanists avoiding jo and ja nai in ordinary speech, and instead using prepositions like fau - in the event of..., va'o - under conditions..., se ja'e - results from ... happening, or ni'i - logically caused by....

Better examples of using them in ordinary speech:

ja gi nai je gi ro da poi badna zo'u da pelxu gi mi citka lo badna gi mi citka lo pelxu
If everything for every banana it is true that it is yellow, and I am eating a banana, then I am eating something yellow.
ja gi nai je gi ro pipybanfi cu crino gi la .kermit. cu pipybanfi gi la .kermit. cu crino
If every frog is green, and Kermit is a frog, then Kermit is green.

Non-logical conjunctions

lo jisra joi lo djacu - the juice and the water, considered together

This is the same as lo gunma be lo jisra je lo djacu.

djacu = x1 is some water
jisra = x1 is some juice

Not all English sentences containing and are like this, though. Firstly, sentences like I had a bath and washed my hair are structurally different and will be dealt with later on. Secondly, I visited Ranjeet and Jasmine is slightly different from I visited Ranjeet and I visited Jasmine. In this case, you probably want to say that you visited Ranjeet-and-Jasmine as a unit on one occasion — not that you visited Ranjeet and Jasmine on (potentially) different occasions (It is true that I visited Ranjeet, and it is true that I visited Jasmine.) In this case you don't want je (which is true but potentially misleading), but joi, which means ‘in a mass with’. So what you have is

mi pu vitke la .ranjit. joi la .jasmin.
I past visit Ranjeet in-a-mass-with Jasmine.
I visited Ranjeet and Jasmine (together).

This is just like the difference between lo ci gerku and loi ci gerku which we looked at in Lesson 4 — considering the three dogs as individuals, or as a mass. Incidentally, it is not just Lojban which makes this distinction; Turkish, for example, would use ile (‘with’) rather than ve (‘and’) for joi here.

By non-logical conjunctions, we mean that the truth of the combined terms does not depend on the truth of the individual components. It may not be true that la .kris. cu bevri lo pipno. Chris carries the piano, or la .pat. cu bevri lo pipno Pat carries the piano, for example (to revisit an example from Lesson 4), even if it is true that la .kris. joi la .pat. cu bevri lo pipno Chris and Pat carry the piano.

Lojban has several other non-logical conjunctions; we'll cover the most frequently used ones:

  • fa'u carries the meaning of respectively: it relates pairs of sumti cross-wise. If I were to say
la .alis. je la .jasmin. cu tavla la .djang. je la .ranjit.

that means that both Alice and Jasmine talk to both Zhang and Ranjeet. If I want to say that Alice only talked to Zhang, and Jasmine only to Ranjeet (i.e. Alice and Jasmine talked to Zhang and Ranjeet, respectively'), a logical conjunction is not useful. Instead, I would use fa'u to connect both pairs of sumti:

la .alis. fa'u la .jasmin. cu tavla la .djang. fa'u la .ranjit.

Alice, cross-wise with Jasmine, talks to Zhang, cross-wise with Ranjeet.

  • If you're talking about a range, you use bi'i to describe the range between the first thing and the second thing; so it corresponds to English between. If you want to say I dropped my pencil somewhere between the office and the bar, you would describe the location somewhere between the office and the bar as lo briju ku bi'i lo barja. The whole sentence would come out as:
mi falcru lemi pinsi vi lo briju bi'i lo barja
This selma'o, BIhI, like selma'o JOI to which all non-logical conjunctions belong, can join both sumti and selbri.


  • If the order of the things defining the range matters, you use bi'o. This corresponds to from... to... in English (though between covers both ordered and unordered intervals.) For example, from 1 PM to 2 PM is an interval lasting an hour; but from 2 PM to 1 PM would normally be interpreted as a 23-hour interval (1 pm the following day), since times in English are assumed to be presented in order. Lojban follows suit with li pavo lo'o bi'o li paci as a 23-hour interval. If I said li pavo lo'o bi'i li paci, the order of the two times would not matter at all; so I could still be talking about a one-hour interval instead.
Tip: The selma'o BIhI needs all sumti terminated before it, not just normal sumti with lo or lo. Since numbers are also sumti, you have to use the terminator corresponding to li, which is lo'o.


Note: You can use non-logical conjunctions in forethought mode, too: the forethought conjunction is the non-logical conjunction followed by gi. So the forethought version of la .kris. joi la .pat. is joi gi la .kris. gi la .pat.


The word "logical" in "logical connective" refers to the association a logical connective has with a truth function. Not all useful connectives can be defined through a truth function, however, and so there are other connectives beside the logical ones.
The meaning of a logical connective is defined the same as two different bridi connected with that logical connective. For instance, mi nitcu do .a la .djan. is defined to be equivalent to mi nitcu do .i ja mi nitcu la .djan.. This definition is useful to bear in mind, because it implies that sometimes, sumti cannot be connected with logical connectives without chaning the meaning. Consider the sentence: "Jack and Joe wrote this play." One attempt at a translation would be: ti draci fi la .djak. e la .djous.
draci x1 is a drama/play about x2 by writer/dramatist x3 for audience x4 with actors x5
The problem with this translation is that it means ti draci la .djak. ije ti draci la .djous., which is not really true. Neither Jack nor Joe wrote it, they did so together. What we want here is of course a mass, and some way to join Jack and Joe in one mass. This has little to do with a truth function so we must use a non-logical connective, which are of selma'o JOI. We'll return to this Jack and Joe-problem in a little - first: Four of the known JOI:
The Lojban connective: ce ce'o joi jo'u
Joins sumti and forms a: set sequence mass group of individuals
The functions of these words are simple: lo'i remna jo'u lo'i gerku considers both the set of

Raising

tu'a - convert noun to vague abstraction involving the noun. Equivalent to lo nu + the noun + co'e vau

vau needed?

So you can translate tu'a as some abstraction associated with..., or more colloquially, some stuff about.... tu'a is easily the most popular way of dealing with abstractions you wish weren't there in Lojban; Lojban sentences using it come out fairly similar to the natural language sentences without abstractions that we're used to seeing.

The other function of jai is easier to explain. It simply converts the selbri such that the noun in the x1 gets a tu'a in front of it (ko'a jai broda = tu'a ko'a broda). In other words, it converts the selbri in a way such that it builds an elliptical abstraction from the noun in the x1, and then fills x1 with the abstraction instead of the actual noun. Again, the original noun place is accessible by fai. A very active Lojban IRC-user often says le gerku pe do jai se stidi mi, to use a random example of a noun in x1. What's he say?

stidi = x1 inspires/suggests x2 in/to x3

Answer: “I suggest (something about) your dog.”

When looking up words in the Lojban dictionary, you may have already noticed that, where languages like English have people or things as subjects and objects, Lojban often uses abstractions instead as gismu places. For example, in English, you say that someone is interesting, or something is interesting. In Lojban, you aren't really meant to say either. E.g.:

cinri - x1 (abstraction) interests/is interesting to x2; x2 is interested in x1

In other words, as far as Lojban is concerned, it's not things or people that are interesting, but actions or properties involving those things or people. For example, Jasmine cannot be said to be interesting simply by virtue of being Jasmine; the way Lojban puts it, it's the things Jasmine does (or is) that are interesting — the way she talks about British sitcoms, her choice of headgear, her tendency to break into '80s songs after she's had a few drinks.

fenki - x1 (action/event) is crazy/insane/mad/frantic/in a frenzy (one sense) by standard x2

People are called crazy. Only occasionally are actions also called crazy. In Lojban, however, the verb fenki describes events (or actions, which is the same). In other words, as far as Lojban is concerned, craziness lies in actions, not in people; a crazy person is by definition someone who does crazy actions.

Note: This means that someone suffering from the particular forms of mental illness loosely called ‘crazy’ wouldn't be called fenki in Lojban — since their condition is not primarily a matter of socially unacceptable actions — but rather menli bilma - mentally ill.

What if you want to say Jasmine is interesting, but you don't care or just unsure what verb to use? I should at least be able to say something like Jasmine {doing some stuff I'm not listing here} is interesting, or Some things about Jasmine are interesting. In other words, I have to say

lo nu la .jasmin. cu co'e cu cinri

or

tu'a la .jasmin. cu cinri

Saying Zhang is crazy (or berserk, probably a closer translation of fenki), I don't have to enumerate the various wacky stunts he has pulled over the years. I can simply say that some stuff about Zhang is crazy, which in Lojban comes out as

lo nu la .djang. cu co'e cu fenki

The value of co'e could be

  • dasni lo zirpu - wears purple
  • dansu la zgikrfanki jipci - dances the Funky Chicken
  • tavla bau la .lojban. - speaks Lojban

or whatever; we're just not bothering to name it here.

So the usual Lojban for Jasmine is interesting is

tu'a la .jasmin. cu cinri

and the usual Lojban for Kevin is crazy is

tu'a la .kevin. cu fenki

Lojbanists noticed how linguists have been analyzing these concepts in natural languages, and how they were coming up with their own versions of verbs. Often, what was a noun in one part of the sentence, and a verb in another part, were brought together and considered to be underlyingly part of the same abstraction noun.

A good example is the phrase I am difficult to annoy in English. At first sight, you might think that I is a noun of difficult. And grammatically it is: it's the subject. But logically it isn't: what we're describing as difficult is not me. We can't say:

  • Who is difficult?
  • Me (to annoy).

What's actually going on is that, underlyingly, what is difficult is to annoy me: the action of getting me annoyed is what is hard to achieve — not me! This is why English also allows you to say It is difficult to annoy me, and (if you squint a little) To annoy me is difficult. And sure enough, Lojban expresses this concept according to that ‘underlying’ form:

lo nu fanza mi cu nandu
The event of annoying me is difficult

So why did English pull that weird switcheroo with I am difficult to annoy? Basically, because when we talk, we aren't concentrating in our minds on intangible abstractions like the event of annoying me, let alone the state of Jasmine having certain unspecified properties. Instead, we run little stories in our head, with heroes and villains: concrete heroes and villains — people, for the most part. And as it happens, we make the subjects of our sentences be the heroes and villains we're concentrating on. (That's what a subject's ultimate job is: to present what we're concentrating on).

So by pulling a switcheroo like that, we're not talking about abstractions and events any more; the subject of the sentence is now our perennially favorite subject — namely me: it's me that is difficult to annoy. (Yes, it is all about me...) This process is called in linguistics raising, because it raises concrete subjects (and objects) we want to talk about, out of the haziness of an abstraction noun (or ‘clausal argument’, to use English logical terminology).

Another case is saying

Jasmine is interesting and beautiful.

We could say

tu'a la jasmin cu cinri .i la jasmin cu melbi

as melbi requires a noun as it's first (x1) place.

But that would be bulky. We have another method

tu'a la jasmin cu cinri = la jasmin cu jai cinri

So jai extracts a noun from the verb place containing an event. Thus, it's very useful:

la jasmin cu jai cinri vau je melbi
Jasmine is interesting and beautiful.

Now both verbs have the same noun in common.

Another example. How do I say that someone is a cheat, or a deceiver? The verb for "to deceive" is:

tcica - x1 (event/experience) misleads/deceives/dupes/fools/cheats/tricks x2 into x3 (event/state)

So lo tcica is a trick, not a trickster; a deception, and not a deceiver. To say that someone is a trickster or a deceiver, we need to use tu'a: tu'a zo'e tcica. But you can't put lo in front of tu'a da: the deceiver has to be the x1 of some verb, in order to get their own noun. So we say lo jai tcica.

jai changes the place structure of the verb involved. This works just like se changing the place structure of it's verb, swapping its first and second place. If we put jai in front of a verb, its x1 place changes from an event, to any noun contained within the event. Let's try this with a few sentences:

  • lo nu la .kevin. cu dasni loi zirpu cu fenki
  • la .kevin. cu jai fenki
  • lo nu la .jasmin. cu co'e cu cinri
  • la .jasmin. cu jai cinri
  • tu'a la .jasmin. cu tcica la .alis.
  • la .jasmin. cu jai tcica la .alis.
  • lo nu fanza mi cu nandu
  • mi jai nandu

You'll notice that, with these new place structures, the Lojban phrases sound pretty much like their English equivalents. For example,

la .jasmin. cu jai cinri
Jasmine is interesting.
la .jasmin. cu jai tcica la .alis.
Jasmine deceives Alice.

We can now do with jai those things we couldn't before. The Lojban for Jasmine is interesting and beautiful, for example, is

la .jasmin. cu jai cinri vau je melbi

That's because Jasmine goes in the x1 place of jai cinri, just as it goes into the x1 place of melbi. And if I want to make a noun meaning ‘deceiver’ or ‘trickster’, I can use jai to do it:

tu'a la .jasmin. cu tcicala .jasmin. cu jai tcicalo jai tcica

However, mi jai nandu does not correspond to I am difficult to annoy. In switching a concrete noun for the original x1 — the event that was difficult — we have lost the abstraction itself: there is nothing in mi jai nandu that means to annoy. But not to worry: Lojban allows you to keep the original event in the verb phrase by preceding it with fai. fai is a place tag like fa and fe. It effectively adds a new place to the verb phrase. So I am difficult to annoy is matched almost word-for-word by the Lojban sentence

mi jai nandu fai lo ka fanza mi

And we can apply this pattern further afield; for example, the book took three months to write is in Lojban properly

lo nu finti lo cukta cu masti li ci
To write the book had a month-duration of three.'

???what is raising? Raising allows the slightly more familiar-looking

lo cukta cu jai masti li ci fai lo ka se finti

jai allows you to talk about things in a way that is in many ways more natural.

Referring to main verb phrase

I use BAI in such a way that whenever their source predicate has a noun place that could integrate the main bridi, then they will do so (I explained the same thing in the {to'e ri'a} thread).

{du'i} is one such BAI. I define it thus:

broda du'i X ~= X dunli lo su'u broda

So, I would say, for example:

(1) do bajra du'i lo nu lo tirxu cu bajra "Your run is like that of a tiger."

It is seldom the case that a concrete object is {dunli} to an abstraction, although it is not impossible. But usually, replacing the X in the above equation with a concrete object will create "non-sense". But we can use {tu'a X} when we don't want to specify the full abstraction, or when it's obvious.

(2) do bajra du'i tu'a lo tirxu "You run like a tiger [runs]"

The expansion of this {tu'a X} is almost always {lo nu X no'a}, which corresponds to the English in []-brackets in (2).

.i tu'a zo du'i pu smuvanbi .i lo do kanla cu dunli lo nu lo lunra cu te gusni lo nicte kei lo ka ma kau ni melbi

Numbers

So the numbers from one to nine are as follows:

  1. pa
  2. re
  3. ci
  4. vo
  5. mu
  6. xa
  7. ze
  8. bi
  9. so

Notice that the numbers repeat the vowels AEIOU. zero is no (think yes, we have no bananas).

pano 10
zebi 78
xanoci 603
vomusore 4,592

4,592 has a comma in it (or a full stop in some languages, just to make things confusing). We can't use a comma in Lojban, because that means separate these two syllables (as we saw in Lesson 1 with Lojbanized names like .zo,is. for Zoe). What we say instead is ki'o. We don't have to use ki'o, but it can make things clearer. So 4,592 can also be read as vo ki'o musore. ki'o also has the advantage that if the following digits are all zeroes, we don't need to say them, so 3,000 is ci ki'o. You can remember ki'o easily if you think of kilo — a thousand. (The similarity is not coincidental).

Just as we have a word for a comma, we also have one for a decimal point: pi. So 5.3 is mupici. In fact, pi is not always decimal; it's the point for whatever number base you're using. But that's a more advanced topic.

Tip: Don't get this mixed up with the number pi (π): 3.14159..., which has its own word in Lojban: pai — oddly enough.

When you want to talk about numbers as sumti — that is to say, as things in and of themselves — you need to put an article in front of them. But that article cannot be la, and for reasons which hopefully will become clear soon, it cannot be lo either. In front of numbers, Lojban uses the article li. So li pareci means ‘the number one hundred and twenty three’. ‘One, two, three’, on the other hand, would be li pa li re li ci: each li introduces a brand new number.

Proportions

ci lo vo mensi pe mi cu nelci la .rikis.martin.
Three of my four sisters like Ricky Martin.

This states two facts:

  1. I have four sisters.
  2. Three of them like Ricky Martin.

But it doesn't actually state that my fourth sister hates him — she may be indifferent to him, or never have heard of him.

One way out of this problem is to use fi'u, which is like the Lojban slash sign. So two out of every three people is really 2/3 of people, or re fi'u ci lo prenu.

Of course, this is actually a fraction, and fractions have decimal equivalents. You could also say 0.75 - pi ze mu lo prenu.

WRONG! Yes, that's our new friend loi in that sentence. If I had said re fi'u ci lo prenu, that would have to be understood in the same way as re lo prenu or ci lo prenu (i.e. as a count of individuals), and I would have ended up talking about two thirds of a person. In most cultures, chopping up persons into thirds is not considered acceptable behavior even for pollsters or advertisers. On the other hand, chopping up populations into thirds is perfectly acceptable; and that's what lo prenu is. (A population, I mean, not an acceptable. Though on second thought...)

Here are some more proportions:

mi tcica pimu loi prenu
I fooled half of the people (treating the people as a mass, or population)
mi tcica pafi'ure lei prenu
I fooled one out of two people (which means exactly the same thing)
mi tcica pa lo re mlatu
I fooled one out of the two cats (treating the cats as individuals)
mi se slabu vopano lo pacivore gismu
I am familiar with 410 out of the 1342 (existing) gismu

mi na nelci ro lo gerku
It's not true that I like all dogs.

(This is not the same as I don't like any dogs, which would be mi nelci no lo gerku. There are other ways of saying this, but we haven't got enough grammar under our belt yet).

Summary

la that named
le that one previously mentioned or known from context or this speech
lo that which is/does
li the number

(lu is not an article, it's a quotation mark!)

la'e the referent of (not really an article, as it takes a full sumti or pro-sumti, as in la'edi'u, what the last sentence refers to, as opposed to di'u, the actual words of the last sentence).
le'e the typical previously mentioned or known from context
lo'e the typical
lai the mass named
lei the mass previously mentioned or known from context
loi the mass which is/does
la'i the set named
le'i the set previously mentioned or known from context
lo'i the set which is/does

(Sets turn out to be pretty useful in Lojban, as we'll see towards the end of this course).

We also looked briefly at lu'o, which turns a set into a mass, and lu'a, which turns a mass into a set of individuals (‘group’ and ‘ungroup’). Strictly speaking, these aren't articles, though.

Build more precise words from basic

Noun converters:

  • lo'i flora = bouquet
  • lo'i taxfu = suit of clothes
  • loi prenu = crowd

Adverbs:

  • ve'i cmana = hill
  • ve'u xamsi = ocean

Suffixes:

  • plisytricu = apple-tree
  • plisygrute = apple (fruit)

-pre:

  • vacrypre = airman
  • brijypre = office worker
  • ladrypre = milkman
  • flalypre = lawyer

-stu - place:

  • citkystu = dining-room
  • vikmystu = toilet
  • vitkystu =guestroom
  • dracystu = drama theater
  • lumcystu = bathroom

-sro - store:

  • cidjysro = fridge, foodstore

-tci - instrument:

  • katnytci = chisel
  • ciskytci = typewriter, keyboard
  • prinytci = printer

lujvo and Deep Gismu Structure

jvajvo or regular compound words don't differ much from other verbs. The point in having jvajvo is to quickly create new meanings from existing ones. For example, if kalri means to be open then kalrygau means to open (something)

The rule is that we drop the last vowel from the gismu (root verb), add y and then add the suffix -gau.

mi kalrygau lo canko = I open the window.

mi gasnu lo nu lo canko cu kalri

gau mi lo canko cu kalri

mi kargau lo canko

All four sentences mean the same. kargau is a synonym to kalrygau. What style you choose depends only on you. Thus you can safely speak without any jvajvo always resorting to prepositions like gau or full verbs like gasnu.

There is no need in memorizing kalrygau separately as it is easily derived from kalri once you know what the suffix -gau means.

Here are some most important suffixes:

  • -gau means "to bring about, to be the agent of (some event)" just like we have the preposition gau. Thus compound verbs with -gau describe animated being performing actions like with me opening the window in the last example.
  • -bi'o means "to turn"
  • -se'e means "to cause (some event)" and describes events causing other events. It corresponds to the preposition seja'e and the verb se jalge.
  • -dji means "to want", corresponds to djica
  • -mau means "to exceed", corresponds to zmadu, it creates comparative verbs. xamgu = good, xamgymau = better.
  • -me'a means "to be less", corresponds to mleca, it creates comparative verbs.
  • rai denotes extreme value. xamgyrai = best.

We also have prefixes that are put to the beginning of verbs.

  • nal- means not and corresponds to na'e
  • tol- means anti- and corresponds to to'e.
  • sel- corresponds to se, tel or ter to te, vel to ve and xel to xe.
    selycpacu = se spacu = to be received
    terymukti = telymukti = te mukti = to intend to (do something)
  • zil- corresponds to zi'o and removes the first place from the verb
  • mal- corresponds to mabla and gives a bad connotation of the verb
  • zan- corresponds to mabla and gives a bad connotation of the verb
  • nun- corresponds to nu and turns the verb into an event verb

If a verb starts with a vowel then you put a prefix, then y' and then the verb:

akti = to be active, in operation
aktygau = to activate
toly'aktygau = to deactivate

The y can be dropped if there is only one consonant from each side and both of them are either voiced or voiceless

Omitting y

Sometimes you can omit y when creating a lujvo. The condition is the following:

  1. The consonant from each side of the y is not adjacent to other consonants
  2. They are no the same consonant
  3. If one of the consonants is voiced (b,d,g,z,j,v) and the other one is voiceless (p,t,k,s,c,f) then y can't be omitted.

Examples:

  1. In kalrygau y can't be omitted as the rule 1 would be broken. There are two consonants to the left of y: lr. However, in nunycasnu n to the left and c to the right are adjacent to vowels, not to consonants so we can remove y getting nuncasnu

Negative lujvo

Note: na has its own rafsi, nar; but na'e is more useful in creating new words. na'e in a selbri still indicates an existing kind of relationship, which you would want to describe with a single lujvo; while na could mean anything, including non-existence — making it too broad a concept for most uses.

lo naljmi one who does not understand (from jimpe ‘understand’)
lo naljvi a non-competitor (from jivna ‘compete’)
lo nalkri a non-believer/skeptic (from krici ‘believe’)
lo nalyla'e an unlikely event (from lakne ‘be likely’)
lo nalre'a a non-human (from remna ‘be human’)

We can see that nal is like the English non-, but we need to remember that non-sometimes has other meanings or associations that nal does not have. lo naljvi is simply someone who is not taking part in a competition, not a ‘non-contender’ in the sense of someone who competes but doesn't stand a chance of winning. Similarly lo nalre'a is someone who is not a member of the species homo sapiens (e.g. a chimpanzee or Klingon), and cannot be applied to someone who is inhumane or perceived as subhuman in some way.

We can also use nal with sel and its relatives; for example,

lo naltertcu not a purpose/activity for which something is needed; something which has no requirements (from nitcu ‘x1 needs/requires/is dependent on/[wants] necessity x2 for purpose/action/stage of process x3')
lo nalveltu'i an area of disagreement; a controversial issue (from tugni ‘x1 [person] agrees with person(s)/position/side x2 that x3 (du'u) is true about matter x4')
lo nalselzi'e something you are not free to do (from zifre ‘x1 is free/at liberty to do/be x2 (event/state) under conditions x3')
lo nalselsanji something you are unaware of (from sanji ‘x1 is conscious/aware of x2 (object/abstract); x1 discerns/recognizes x2 (object/abstract)'; this gismu has no suitable short combining form)
lo nalselse'i someone who lacks a self/ego; an enlightened person according to Hindu/Buddhist philosophy (from sevzi ‘x1 is a self/ego/id/identity-image of x2')

As you'll have guessed, the companions of na'e, namely to'e and no'e, have rafsi of their own: tol- and nor-, respectively. So ‘disinterested’, ‘uninterested’ and ‘bored’ in Lojban are norselci'i, nalselci'i and tolselci'i.

lujvo can be much more interesting than this; interesting enough, in fact, that we won't be covering them any further here. You can make lujvo out of pretty much any tanru you can devise; this is the main way to introduce ‘new words’ into Lojban. But to make the lujvo you come up with work, you need some background knowledge:

  • how to make sure rafsi in a word stick together unambiguously in Lojban grammar (The Complete Lojban Language, Chapter 4.5–4.6, 4.10–4.12).
  • how to make sure the gismu inside your tanru group together properly (The Complete Lojban Language, Chapter 5).
  • how to derive the place structure of the lujvo from the place structures of the gismu that make it up (The Complete Lojban Language, Chapter 12).

It's worth your while to look into these issues if you'll be using the language seriously, and especially if you'll be writing in it. (lujvo are easier to deal with while writing than while speaking, because you have the time to reflect on how you'll be creating your new word.) At this stage, though, you don't need to go into all that just yet.

(See also: http://selpahi.weebly.com/17/post/2014/02/how-to-get-rid-of-formal-rafsi.html )

All of the following affixes produce a regularly derivable brivla with at most one added place. Some will add zero places, some one place.

(Note: The option to enter non-derivable definitions for words using a suffix should not be taken away entirely, as it would mean a restriction in what words we can make up. I would suggest to maintain at least a bit of freedom just in case. Still, in the vast majority of cases, the suffixes should yield easy to understand derivations. Extra note: In contrast, words that don't use any catalogued affixes are free to mean whatever their inventors choose. As long as it is morphologically a brivla, it can mean whatever. Lujvo/Gismu/Fu'ivla are no longer regarded as mutually exclusive groups, meaning that a "lujvo" shape can have a fu'ivla definition, or a fu'ivla can have gismu shape, etc. Another consequence is that type 3 fu'ivla and most kinds of type 4 fu'ivla have no reason to exist anymore. A word like {tcizbaga "cheeseburger" is a valid borrowing, because it's a brivla, and we no longer have to add annoying clusters to artificially change its type to fu'ivla.)


Prefixes:

These are always placed before a brivla.

Prefixes are preferably CCV- so that we don't have to worry about tosmabru. People are supposed to be able to use them spontaneously and creatively, and ideally tosmabru should not be a possible danger. CV[nlr]- is also tosmabru-safe.

affix source effect example


cfa- (cfari) beginning cfaviska "to catch sight of something" zan- (zabna) positive conn. ... mal- (mabla) negative conn. ran- ? (ranji) continuously ranbajra "run continuously" "keep running" ? (purci) ex- can- (canlu) around an area canbajra "to run around" xru- (xruti) re/back xrudunda "to give back"; xrudarxi "to hit back" pru- (purci) pre-/ahead ... bra- (barda) big cma- (cmalu) small rem- (remna) human

Suffixes:

These are always placed at the end of a brivla.

affix source effect example


-bi (binxo) become xunrybi "to become red" "to blush" -zu (zukte) action crezenzu "to practice" -mu (mukti) motive -ri (rinka) cause tatpyri "tiring" -ja (selja'e) result/outcome glekyja "result in being happy" -gau (gasnu) bring about jungau "tell" -dji (djica) want djunydji "to want to know" -ju (djuno) know klamyju "know that someone goes" -tce (mutce) very -ru (ruble) -mau (zmadu) -me (mleca) -rai (traji) -ze (zenba) -di (jdika) decrease glarydi "to cool down" -cu/ki (cumki) possibility cpacycu/cpacyki "available"/"acquirable" -kre (krefu) re-/again krezbasu "rebuild" -si/su/xu? (simxu) each other pramysi "to love each other" -sa? (simsa) -oid/-like remnysa "humanoid" -lu? (simlu) seem morsylu "to appear dead" -ni? (ckaji jibni) -almost -tu/ci?(tutci) tool/utensil ciskyci/ciskytu "writing utensil" ... -fe (fetsi) feminine baknyfe -na (nakni) masculine baknyna

      ()         group/crowd   remny-CV "a crowd", cinfy-CV "pride of lions"

-sra (srana) -ly/pertaining to cevnysra "godly"


+ ranji, tutci, vasru, stuzi, tcaci, cmima, simsa, troci...; prenu?? ... + diminutive and augmentative suffixes

Also consider "compound affixes" CVCV combining ideas from multiple other words to get a new affix, possible examples include "-bizu" from "-bi'ozu'e" and "-pazu" from "pagzu'e" (to participate).

-sebi (se bilga) obligatory plejysebi "must be paid for" -secu/seni (seltcu/se nitcu) required tcidysecu/tcidyseni "required reading"

lo bi'u nai se cukta cu tcidysecu "The book is required reading."

lo nu nerkla cu plejysecu "You have to pay to go in."

Another idea: A way to derive "easy to X" and "hard to X", e.g. "easy to provoke" or "hard to understand".


Make a list of shapes that are safe from tosmabru no matter what affix gets added. Those word shapes are the most robust and should be preferred over the others. Safe: CVC/CVCV CCVCVCV CCVCV CVC/CV ...

Unsafe: CVCCV ...

For safe zi'evla shapes, adding -CV suffixes is possible in two ways: Either by replacing the final vowel with -y- and adding -CV or by leaving the vowel alone and adding -CV. E.g. {delfinu} -> {delfinyfe}/{delfinufe}. Doing the latter with unsafe brivla shapes causes tosmabru. Doing it with CCVCV (safe) gismu shapes turns them into safe zi'evla shapes (e.g. gerkufe "female dog"), which could cause collisions with already existing zi'evla. Maybe what gets defined first wins.

zi'evla

loan words are a way of introducing new words into Lojban, comparable tolujvo.

You got a brief taste of lujvo in Lesson 8. As we said there, lujvo are the main way of introducing new words — more precisely, new brivla — into Lojban. The most important thing about lujvo is that, as selbri, they are meant to have very well-defined place structures; and there are guidelines in place for deriving them (see The Complete Lojban Language, Chapter 12.) So, particularly when the concept you want to express is ‘verb-like’ (that is, when it's likely to have sumti of its own), lujvo are preferred.

There are some cases, though, when you do have to borrow a word from another language, creating a loan word (called in Lojban a fu'ivla). This can be because the thing you're talking about is very concrete or particular, and/or because the reference is quite culture-specific. In either case, it would be really cumbersome to describe it with a combination of gismu. (For example, how would you come up with a description for brie? Or rock 'n' roll? — which, we should point out, you would have to keep distinct from the later musical genre of rock!)

The problem with borrowing words into Lojban is, Lojban has a quite thorough set-up for working out what the words are in a stream of letters. This means that most words you import into Lojban (once you spell them in Lojban letters) are likely to mean something else already. For example, if I want to bring the ancient name of the river Danube - Istros into Lojban, the last thing I want to do is start saying .istero. That will get analysed as .i stero, which is something like ‘and steradian function’.

The sanctioned way to deal with loan-words (described in more detail in The Complete Lojban Language, Chapter 4.7) is to stick a gismu (minus its final letter) in front of the word, showing what sort of thing the word is; and to put an r (or, if an r is already there, an n) between the gismu and the word. The gismu helps the reader or listener, who has likely never seen this word before, guess what the word might be. This is particularly handy if the source word might be ambiguous between two different meanings. And the combination of gismu minus final vowel, source word (which should start with a consonant, and end with a vowel), and r or n will hopefully produce a cluster of consonants crunchy enough that it cannot be mistaken for another Lojban word or phrase.

Tip: There is no standard consonant to put in front of the word to become a fu'ivla if it starts with a vowel. Two popular choices are x and n. Similarly, there is no set convention on where to get the vowel from, if your word ends in a consonant. In these lessons, we'll just repeat the preceding vowel; e.g. England  gugdrninglanda (from gugde ‘country’).

So what does all this look like in practice? Well, we've already seen curry:

  • take ‘food’, cidj[a];
  • take the word in Lojban garb (starting with a consonant and ending with a vowel), kari;
  • and wedge them together with an r: cidjrkari.

(The consonant cluster is also crunchy enough to be difficult to pronounce; the r is a syllable on its own, and the word should sound something like shidgerrrrrkari).

Loan words (in Lojban, fu'ivla) are still only sporadically used — particularly because, as of this writing at least, there is no Lojban dictionary where a standard list of them can be looked up. The problem of which language to borrow words from is also hard to settle, and the choices made can cause problems of their own. The most international solution for plant and animal names, for example, is Latin, and in particular the Latin of the Linnaean system of classification. But this means that, to come up with a word for ‘catnip’, say, you have to know Latin and your Linnaean taxonomy. (Or, like I did, look it up on the Internet — but you can't normally do that while you're having a conversation.) So fu'ivla are still largely unexplored terrain in Lojban.

Note: That said, you will occasionally see ‘Stage 4' fu'ivla in use. The fu'ivla we've seen are ‘Stage 3'; in Stage 4, you drop the initial ‘crunchy’ rafsi, reasoning that the word should already be well-known or recognisable enough — and making sure that the word still doesn't look like a normal brivla. (For example, The Complete Lojban Language suggests tci'ile for ‘Chile’, instead of gugdrtcile.) Not everyone likes them, so they're not yet all that common, and you'll usually get plenty of warning if someone is using them. P.S.: If you were wondering, by the way: cirlrbri, zgiknroknrolo, zgiknroko.

Other types of anaphora

he and she

Several letters have a defined meaning, a type of thing that it refers to, reliably. By adding a suffix xoi to the right of them we get corresponding pronouns that refer back to the most recent living thing, the most recent person, object, abstract thing and many more. The letters are based on the first letter of some gismu to make them easier to remember. Below is a list of possible referents.

fie fetsi "she", female, feminine
nue nakni "he", male, masculine
pie prenu "he/she/'ey/they", person
rie remna "he/she/'ey/they", human being
jie jmive (gi'e na remna) "it", animate, animal
due dacti (gi'e na jmive) "it", inanimate object
sue stuzi "it", place, location
mie mucti "it", abstract, immaterial, thoughts, ideas

Let us look at some of the amazing things we can do with this.

na go'i .i mi pu no roi zvati sue
No. I have never been there.
fie dunda lo kargu junla nue
She gave him an expensive watch.
mi je lo mensi be mi pu klama sue .i ku'i fie na nelci sue
Me and my sister went there, but she didn't like it.
ko xrugau due lo pu te tolcri be fi do
Put it back where you found it.
mie cinri sidbo
That is an interesting idea.

rixire

To refer to the next-to-last noun, the third-from-last sumti, and so on, ri may be subscripted using the particle xi:

lo smuci .i lo forca .i la rik. cu pilno rixire .i la .alis. cu pilno riximu
A spoon. A fork. Rick uses [repeat next-to-last]. Alice uses [repeat fifth-from-last].

Here rixire, or ri-Number 2, skips la .rik. to reach lo forca. In the same way, riximu, or ri-Number 5, skips la .alis., rixire, la rik., and lo forca to reach lo smuci. As can clearly be seen, this procedure is barely practicable in writing, and would break down totally in speech. That's why we have ra.

ko'a

Now there are plenty of KOhA sumti to go around. In fact, if you've run out of words by getting to ko'u, you can start over again with fo'a, fo'e ... fo'u. There is a problem, though: you have to remember (a) which sumti was assigned to which KOhA word, and (b) to assign the sumti in the first place. There's nothing to say that this will not become commonplace in future Lojban usage. Right now, however, there is a feeling that this is a little too calculated to work spontaneously. And Lojban cannot readily use the little hints natural languages pepper their grammar with (like gender and number), to keep track of who is who.

As a result, yet another strategy has been introduced to refer back to sumti. This strategy dates back from ‘Institute’ Loglan, before Lojban arose in its modern form. (Yes, Lojban has a history and a prehistory. No, we don't really have the time to go into them here.)

Assigning pronouns

If we're telling a story in English, the meaning of, say, she keeps changing. At the moment, it means ‘Alice’, but if Alice's friend Jasmine walks into the bar, she could very well mean start meaning ‘Jasmine’. In Lojban, we can keep on using lo go'i, ri and their relatives, but there is an easier way of dealing with a larger cast of characters.

What we do is assign pro-sumti as and when we need them, using the cmavo goi (which is like the Latin word sive, or the English also known as (aka)). The sumti assigned by goi are a series called KOhA, consisting of ko'a, ko'e, ko'i ... you get the idea?

Note for lawyers (and frustrated non-lawyers): The equivalent in legal documents of goi is henceforth referred to as, and ko'a is something like the party of the first part. Lojban has in fact been proposed as the ideal language for law, where precision is of utmost importance. It would also allow non-lawyers to understand legal documents, which would be something of a miracle.

OK, let's go back to Alice's story. We start by saying

la .alis. goi ko'a klama lo barja

This means that from now on, every time we use ko'a, we mean ‘Alice’. The man she sees can then be ko'e, so we say

.i ko'a zgana lo nanmu goi ko'e

Now every time we use ko'e, it means that particular man, so the full story so far reads:

la .alis. goi ko'a klama lo barja .i ko'a ze'a pinxe loi vanju .i ko'a zgana lo nanmu goi ko'e .i ko'e melbi .i caku ko'e zgana ko'a

(Note how the cus have disappeared: ko'a, like mi, doesn't need them, since it can't join with a selbri to form a new selbri).

Assigning ko'e to lo nanmu is actually better than starting the next sentence with lo nanmu. This is because lo nanmu simply means the thing I have in mind which I call ‘man’, which is not exactly the same as the man (it could, in theory, be something totally different). Some Lojbanists might even say that using lo like this is a bit malglico. (Or at least malrarbau ‘damned natural languages’: lots of languages have definite articles, and Lojban lo is no definite article).

Tip: If you combine ko'a/ko'e/ko'i/ko'o/ko'u with ri/ra/ru, don't count ko'a-type pro-sumti when you're counting back. For example
la .alis. cu rinsa ko'e .i ri cisma
doesn't mean that ko'e (the man, in this context) smiles, but that Alice smiles. Why? Because it is pointless to have a replacing word (anaphor), like ri, replace another replacing word, like ko'e. If you wanted the x1 of cisma to be ko'e, you would have simply said .i ko'e cisma, not .i ri cisma. It works out simpler to keep ri/ra/ru in reserve for more important things.

Let's continue by introducing Alice's friend Jasmine (if people are wondering where I get all these unusual names from, Jasmine is an old Gujarati friend of mine). We continue ....

la .jasmin. goi ko'i mo'ine'i klama .i ko'i rinsa ko'e
Jasmine (henceforth #3), goes into. #3 greets #2.
Jasmine comes in and says hello to the guy.

mo'ine'i is another space ‘tense’. mo'i indicates movement; ne'i means ‘inside’ (from the gismu, nenri). So mo'ine'i corresponds to the English preposition into (while ne'i on its own corresponds to inside or in.) The way Lojban grammar works, mo'ine'i on its own is treated as mo'ine'i ku: a preposition with an omitted sumti. (Remember caku, which is exactly the same. Just as baku means ‘afterwards’ (relative to the here-and-now), mo'ine'i [ku] means something like ‘in(to)wards' — but is nowhere near as weird in Lojban as it is in English).

mo'i is extremely useful, as it allows you to distinguish between location and motion. For example, I ran behind the bar in English is properly speaking ambiguous: are you running while behind the bar, or are you running with your final destination behind the bar? Lojban does not allow that ambiguity: mi bajra ti'a lo barja means the former, while mi bajra mo'i ti'a lo barja means the latter. In the example given above, ne'i klama would mean not that Jasmine comes in (from outside), but that she is going from somewhere to somewhere else, while inside. This kind of ambiguity may pass unnoticed by native English speakers, but speakers of languages which are more precise about direction find it extremely vague (Turkish, for example, has at least three words to translate ‘here’).

bo for linking sentences

bo has another use, which seems separate from selbri grouping: It can also bind a sumtcita to an entire bridi, so that the content of the sumtcita is not a sumti, but the following bridi. This is best explained with an example.
xebni x1 hates x2
mi darxi do .i mu'i bo mi do xebni – "I hit you, with motivation that I hate you." Here the bo binds mu'i to the following bridi.
This is where the technical difference between tense sumtcita and other sumtcita is relevant. You see, when binding a normal sumtcita to a bridi with bo, it means that the following bridi somehow fits into the sumti place of the sumtcita. For the reason of God Knows Why, binding one of the words ba or pu to a bridi has the exact opposite effect. For example, in the bridi mi darxi do .i ba bo do cinjikca, one would assume that the second bridi is somehow filled into the sumti place ofba, meaning that the bridi first uttered took place in the future of the second bridi. That's not the case, however, and the correct translation of that utterance would be "I hit you. Afterwards, you flirt". This weird rule is actually one of the main obstacles to a unification of all sumtcita into one single word class. Another difference is that tense-sumtcita can be elided, even though they apply. This rule makes more sense, since we often can assume bridi is placed in a time and space, but we can't assume that the sumtcita of BAI applies.

More interjections

One final thing: if you want to know how someone feels about something, once again Lojban provides a fill-in-the-slot question word. The word asking the listener to fill in the interjection that best applies is pei. You can fill pei in with any interjection or sei-phrase. So if I ask you

pei can also explicitly ask for NAI or CAI alone, by following a UI cmavo. So a response to

.i .u'ipei do farlu lo pesxu
You fell into the mud! Funny, eh?

could well be ru'e: Kinda... Then again, it could also be naicai: Absolutely not, and I shall thank you never to mention it in my presence again. (Allowing for some latitude in translation..).

Note: selma'o UI4 specifies what ‘part’ of you is feeling the emotion — whether it is a physical, social, mental response, and so on. selma'o UI5 has some ‘left-over’ modifiers; we already saw in passing ga'i, which indicates haughtiness.

Time journeys

How can I say "broda from morning to evening"? I doubt {broda co'a lo cerni co'u lo vanci} would work?

<latro`a> while I would say that, it is indeed wrong, because it says "the event of brodaing ending, which happens at the evening, begins at the morning"

<latro`a> that is, the co'a tag scopes over the co'u tag, while you want them at the same scope level; as far as I know there's nothing that can be done about this with tag syntax

<latro`a> you can, however, say, {lo nu broda cu cfari ca lo cerni gi'e se fanmo ca lo vanci}

<danr> it might be a bit too formal for my purposes. I'm considering {broda fi'o cfari lo cerni fi'o fanmo lo vanci}

<latro`a_> same problem. {fi'o} has the same scope phenomena

<tsani> you can use termsets to give terms equal scope

<tsani> {.i broda co'a ko'a ce'e co'u ko'e}

Vocatives

  • When it isn't your turn to speak, but you want to barge in anyway, you can say ta'a — though it probably won't make anyone any happier that you're interrupting.
  • Finally, to close communication (radio's Over and out!), you can use fe'o. (This is what people actually should be putting at the end of their e-mails; but it's not as well-known a word as co'o)


mekso saske

Likewise, you can attach a sumtcita: le nu darxi kei be gau do: “The event of hitting, which is caused by you”. Note that the presence or absence of kei makes it parse differently: With the fa'orma'o present, be attaches to nu, without the fa'orma'o, it attaches to darxi. So it decides what is being emphasised: Is the hitting, or the event of hitting caused by you? In this specific case, though, that's just about the same thing.

famyma'o Warning!: One could have description nouns inside description nouns, saying le le se cinjikca be mi ku gerku, = le gerku pe le se cinjikca be mi =“the dog of the man I'm flirting with”, but that's not very easy to read (or to understand when spoken), and is often being avoided.

One need also to learn tu'a, since it will make a lot of sentences much easier. It takes a sumti and converts it to another sumti - an elliptical abstraction which has something to do with the first sumti. For example, I could saymi djica lo nu mi citka lo plise, or I could let it be up to context what abstraction about the apple I desire and just say mi djica tu'a lo plise. One always has to guess what abstraction the speaker means by tu'a SUMTI, so it should only be used when context makes it easy to guess. Another example: gasnu “x1 does/brings about x2 (volition not implied)” za'a do gasnu tu'a lo skami - “I see that you make the computer do something”. Officially, tu'a SUMTI is equivalent to le su'u SUMTI co'e. Vague, but useful. There are situations where you cannot use tu'a, even though it would seem suitable. These situations are when I don't want the resulting sumti to be an abstraction, but a concrete sumti. In this case, one can use zo'e pe. tu'a convert sumti to vague abstraction involving the sumti. Equivalent to le su'u SUMTI co'e kei ku

Finally, one kind of sumti can be turned into another by the words of the class LAhE. lu'a -Convert individual(s)/mass/sequence/set to individuals. lu'i - Convert individual(s)/mass/sequence/set to a set. lu'o - Convert individual(s)/mass/sequence/set to mass. vu'i - Convert individual(s)/mass/sequence/set to sequence; the order is not stated. The use of these words is straight-forwardly: Placing them before a sumti of a certain type makes a new sumti of a new type. Notice, though, that as a fourth kind of sumti, a sequence has been introduced. This is not used very often (it doesn't have its own gadri, for instance), but just included here for completion. The last two members of LAhE do not convert between types of sumti, but allows you to speak of a a sumti by only mentioning something which references to it: If one sumti A refers to a sumti B, for instance because sumti A is a title of a book, or a name, or a sentence (which always refer to something, at least a bridi), ‘'la'e SUMTI A refers to sumti B. For instance, ‘'mi nelci la'e di'ufor “I like what you just said” (not mi nelci di'u which just means "I like your previous sentence") or ‘'la'e le cmalu noltru for the book “The Little Prince”, and not some little prince himself. The cmavo ‘'lu'e does the exact reverse – ‘'lu'e SUMTI refers to an object which refers to the sumti. la'e - “the thing referred to by” - extracts a sumti A from a sumti B which refers to A. lu'e - “the thing referring to” - extracts a sumti B from a sumti A, when B refers to A.


In order to clarify that all tenses are relative to the speaker's current position, the wordnau can be used at any time. Another word, ki marks a tense which is then considered the new standard. That will be taught way later. nau updates temporal and spacial frame of reference to the speaker's current 'here and now'.


In the Lojban tense system, all tenses are prepositions, which we have conveniently just made ourselves familiar with. Okay okay, technically, tenses are slightly different from other prepositions, but this difference is almost insignificant, and won't be explained until later. In most aspects they are like all other prepositions; they are terminated by ku, making it explicit that selma'o (the group) PU is terminated by ku.

  • Every time nu starts an abstraction — a phrase nested inside another phrase —kei ends it.
  • Every time an article like lo or loi starts a sumti, ku ends it.
  • Every time a string of numbers starts, boi ends it.
  • Every time a series of sumti follows a selbri, vau ends it.
  • The binding can be terminated with be'o - one instance of be'o for each selbri which has sumti bound by be. To list them: be binds sumti or sumtcita to selbri bei binds a second, third, fourth (etc.) sumti or sumtcita to a selbri be'o ends binding to selbri
  • li'u is one of the few terminators that can almost never be missed out, since that would make everything else that follows part of the quotation.
  • The elidable terminator for sei is “se'u” (of selma'o SEhU); it is rarely needed, except to separate a selbri within the sei comment from an immediately following selbri (or component) outside the comment.

For example, if we see cu in a sentence, we know that what is coming up is a selbri; so the sumti before it must now be over. So we can drop theku. (In fact, that's why cu exists in the first place: the beginning of a verb is a much more important structural break in natural languages than the end of a noun.) If a new sentence is beginning — as signalled by perhaps the most distinctively Lojbanic word, the ‘audible punctuation’ .i — then (have we covered .i yet?) there can be no more sumti from the old sentence; so we drop the vau. In fact, it is only in situations of potential ambiguity, like the sentence we've been looking at, that you'll get terminators appearing in normal Lojban usage at all.

Note: Remember those pesky possessive constructions from Lesson 3, when you couldn't flip lo tamne pe lo ninmu klama the other way around, because it was ambiguous? All you need is ku to resolve that ambiguity: lo lo ninmu klama ku tamne means 'the woman traveller's cousin', and lo lo ninmu ku klama tamne means ‘the woman's traveller cousin.’ Still, most Lojbanists think the flip-around is not worth the hassle of inserting that bothersome ku, so you rarely see it used when the ‘possessor’ sumti is not a one-word sumti.

(ping!)

cmene

The final point is stress. As we've seen, Lojban words are stressed on the penultimate syllable, and if a name has different stress, we use capital letters. This means that the English and French names Robert come out differently in Lojban: the English name is.robyt. in UK English, or .rab,rt. in some American dialects, but the French is.roBER. .

To give an idea of how all this works, here are some names of famous people in their own language and in Lojban.

English
Margaret Thatcher .magryt.tatcys.

(no th in Lojban because most people around the world can't say it!)

Mick Jagger .mik.djagys.
French
Napoleon Bonaparte .napole,ON.bonaPART.
Juliette Binoche .juLIET.binOC.
Chinese
Laozi .laudz.
Mao Zedong .maudzyDYNG.
Turkish
Mustafa Kemal .MUStafas.keMAL.
Erkin Koray .erkin.korais.
German
Friedrich Nietzsche .fridrix.nitcys.
Clara Schumann .klaras.cuman.
Spanish
Isabel Allende .izaBEL.aiendes.
Che Guevara .tcegevaras.

tanru

The default grouping in Lojban is leftwards. This means that, if you have three things connected together in Lojban, the first two go together before you join in the third. For example, la .jasmin. je la .alis. jonai la .ranjit means not Jasmine and either Alice or Ranjeet, but Either Jasmine and Alice, or Ranjeet.

Does the distinction matter? Depends on your background; programmers, for example, are often driven to distraction in making sure their logical connectives work out in the right order (usually by copious use of brackets.) But there is often a real difference in meaning; the first interpretation given above describes a couple, for example, but the second doesn't.

melbi cmalu ractu kevna

Tatoeba-like. Examples of diff.words

Note: To assume that Lojban works like English in general is a sin Lojbanists are ever on the alert for. It is enough of a community obsession that the Lojban word for it —malglico - damned English — routinely turns up in the English of Lojbanists, even when they're not talking about Lojban. In this instance, it is malglico to assume that ninmu refers to an adult.

This is very Lojbanic — the English word good on its own is so vague as to be almost meaningless. It is also slightly malglico to put a person in the x1 place, which is normally filled by an object, state or event. For ‘morally good’ you would usually use vrude ‘virtuous’.

tu cipni la .serinus.serinus.kanarias.
That is a bird of species 'Serinus serinus canaria'.
mi pritu lo nixli lo nanla.
I'm to the right of the girl, facing the boy.
la .pybysys. cu tivni la .niksyn.in.tcainas. la .kycy,etys. la .telis.
PBS (the American Public Broadcasting Service) televises 'Nixon in China' (an opera) through KCET (the Los Angeles PBS affiliate) to Telly (a pet name for a particular television) (!).
do klama la .uacintyn. la .losandjeles. la .cikagos. la .amtrak.
Julia travels to Washington from Los Angeles via Chicago on Amtrak (the American inter-city train network).

How many sumti can a selbri describe? The number depends on the place structure of the word we use for the selbri. (There are ways of tagging on extra sumti, which we'll cover in later lessons). See jutsi!

Space

Remember the word ti? This is part of a series ti, ta, tu, meaning roughly ‘this’, ‘that’ and ‘that over there.’ If we're talking about places rather than things, we say vi, va, vu, meaning roughly ‘here’, ‘there’ and ‘yonder’ or ‘way over there’. Again, this is determined by the thing you're talking about. If you're telling a doctor where you feel pain, ti might be the end of your toe, while if you're talking about astronomy, ti could be the solar system. We can therefore say

There is a whole class of cmavo that work like ri'u, and they are called FAhA-type cmavo, so named after a (somewhat non-representative) member of their class, fa'a (in the direction of). These include to'o (away from), zo'i (to the same side as), zu'a (to the left of), ne'a (next to), ne'i (within) and so on. (Again, all the space cmavo are explained in Chapter 10 of The Complete Lojban Language).

Note: FAhA cmavo indicate direction, but not motion toward that direction. There is a separate cmavo for that; see Lesson 7.

Getting back to daily speech, these time and space cmavo are very useful for questions. ca ma is ‘simultaneous with what?', or in other words, ‘when?' (a simpler alternative to ti'u or di'e). Similarly, vi ma means ‘at the location of what?', or ‘where?'

Restrictive and non-restrictive relative clauses

Combine the following pairs of sentences into single sentences. In each case, make the second sentence a relative clause modifying the underlined noun in the first sentence. The highlighted noun in the second sentence is the same as that in the first, and will turn into ke'a. You may omit ke'a if possible. Example:

.i mi viska lo botpi .i lo botpi cu culno
.i mi viska lo botpi poi culno

Watch out for vau which you may have to insert.

.i lo ninmu cu dunda lo cifnu lo nanmu .i lo nanmu cu citka loi cidjrkari
.i lo ninmu cu dunda lo cifnu lo nanmu .i lo cifnu cu kakne loka citka
.i lo ninmu cu dunda lo cifnu lo nanmu .i mi pu viska lo ninmu vi lo barja
.i lo ninmu cu dunda lo cifnu lo nanmu .i lonu mi viska lo ninmu cu nandu
.i mi viska va lo barja lo ninmu .i mi klama lo barja lo briju
.i ca lonu mi klama lo barja lo briju kei mi penmi lo nanmu .i lo barja cu snanu lo briju
.i mi viska lo kansa be lo ninmu .i lo ninmu cu dunda lo cifnu lo kansa be lo ninmu
.i mi kakne lo ka citka lo cidjrkari .i lenu citka loi cidjrkari cu nandu
.i lo ninmu cu dunda lo cifnu lo nanmu .i lo nanmu cu citka loi cidjrkari
.i lo ninmu cu dunda lo cifnu lo nanmu .i lo cifnu cu kakne loka citka
.i lo ninmu cu dunda lo cifnu lo nanmu .i mi pu viska lo ninmu vi lo barja
.i lo ninmu cu dunda lo cifnu lo nanmu .i lonu mi viska lo ninmu cu nandu
.i mi viska va lo barja lo ninmu .i mi klama lo barja lo briju
.i ca lonu mi klama lo barja lo briju kei mi penmi lo nanmu .i lo barja cu snanu lo briju
.i mi viska lo kansa be lo ninmu .i lo ninmu cu dunda lo cifnu lo kansa be lo ninmu
.i mi kakne lo ka citka lo cidjrkari .i lenu citka loi cidjrkari cu nandu

swadesh

Nouns animal ashes bark (of a tree) belly bird blood bone child (young) cloud day (not night) dog dust ear earth (soil) egg eye fat (substance) father feather (large) fire fish flower fog foot fruit grass guts hair hand head heart hold (inhand) husband ice lake leg liver long louse man (male) meat (flesh) moon mother mountain mouth name narrow neck night nose person river road root rope salt sand sea (ocean) skin (ofperson) sky smoke snake snow star stick (of wood) stone sun tail tongue tooth (front) tree water wife wind (breeze) wing woman woods worm year

Verbs to blow (wind) to breathe to burn (intransitive) to come to count to cut to die to dig to drink to eat to fall (drop)far to fear to bite to fight to float to flow to fly to freeze to give to hear to hit to hunt (game) to kill to know (facts) to laugh to lie (onside) to live to play to pull to push to rain to rub to say to scratch (itch) to see to seed to sew to sing to sit to sleep to smell (perceive odor) to spit to split to squeeze to stab (or stick) to stand to suck to swell to swim to think to throw to tie to turn (veer) to vomit to walk to wash to wipe

Adjectives back bad big cold (weather) dirty dry (substance) dull (knife) few good heavy here many near new old other red right (correct) right (hand) rotten (log) sharp (knife) short small smooth straight thick thin warm (weather) wet wide Pronouns he I we they thou/you Conjunctions and because if with (accompanying) Prepositions at in Numbers five four one three two Question how what when where who Particles Not Colours black white yellow

  • bilgygau, gau ko'a bilga

Basic vocab

Should be completely inside the course.

  1. прибывать, zukte, санки
    ralte, ponse, предоставлять
  2. fasnu
  3. направлять
  4. Sensory: viska - viskygau - viskyzgana, sance, tinju'i, tirnyzgana
  5. sisku - facki
  6. изволять
  7. ценить
  8. denpa
  9. stali
  10. наречия
  11. более
  12. po'o
  13. mutce
  14. ли
  15. Союзы
    • что; чтобы
    • если, если бы
    • то; же; итак; следовательно,
    • чем; как (при сравнении)
    • или; либо
  16. fa'a seka'a, bu'u, ne'i, sene'i, fa'a, to'o
  17. перед
  18. seba'i ? вместо
  19. вокруг / приблизительно (сколько)
  20. to'o
  21. ze'a
  22. до
  23. между
  24. против
  25. ka'ai, sepi'o - с
  26. по, согласно
  27. посредством
  28. для ; чтобы / за/на – «срок, сумма»
  29. позади ; после
  30. pe - about
  31. cau
  32. над
  33. на
  34. do'e - не имеет определённого значения / все остальные предлоги.
  35. mi, do, ra, mi'ai, do'o, mi, do, za'u ra мы / вы / они
  36. lo drata
  37. lo nei (self)
  38. неопределённо-личное местоимение
  39. 1- who, what, which
  40. 2- where, where to, whose, how
  41. 3- why, for how long, for how many
  42. После успеха с вопросительными запоминать группу утвердительных:
  43. этот, это, такое
  44. Дальше запоминается всё очень легко и быстро. Но оптимальные порядок с моей точки зрения:
  45. da, lo se cuxna be do, ro da, no da
  46. Приставки: anti-/co'a /za'ure'u/no'e/to'e
  47. Суффиксы:
  48. Покупки / Коммерция, банки и торговля
  49. judri
  50. cuxna
  51. заказывать товар/услугу
  52. предлагать товар/услугу
  53. количество
  54. sumji
  55. счёт бухг.
  56. papri, banxe karda лист / что-то плоское на чём можно писать(Бумага) или может быть что-то записано(Банковская карта, Страница Книги или Сайта)
  57. linto - tilju
  58. vikmi - выделять из организма
  59. sarcu
  60. jinsa - toljinsa, lumci / умывальня / умывальник / стиральная машина
  61. Числа и измерения: часть, многий, число
  62. Время дня, года: год, час, момент / моментально – immediately
  63. прояснить(сделать ярче)
  64. danlu: #cinki насекомое, #ползать,
  65. censa, cevni святой – табу,святое место,кошерная еда
  66. течь / жидкость / плавать
  67. дышать
  68. жить, моя жизнь
  69. tsali, ruble
  70. sepli
  71. jbena - рождать
  72. pruxi
  73. sligu/ranti/gapci
  74. barda/cmalu
  75. coi co'o ki'e je'e u'u
  76. sinma
  77. xamgu / tugni do ...
  78. извинять - je'e
  79. ckire, curmi
  80. cusku lo xajmi
  81. Общее, прочее и невошедшее в пред.группы
  82. возраст / совершеннолетний
  83. sraji clani
  84. darxi
  85. красивый
  86. продолжаться
  87. plana
  88. dinju - zdani
  89. sipna
  90. pacna - убеждать, доказывать
  91. оканчивать
  92. mlana
  93. tarmi
  94. tasmi
  95. vlipa - сильный
  96. закрывать (отверстие)
  97. сумасшедший
  98. весёлый
  99. жестикулировать
  100. слава
  101. помогать
  102. впадина
  103. начинать
  104. понимать
  105. сохранять
  106. страна
  107. читать
  108. связывать
  109. язык общения
  110. длинный
  111. играть
  112. светить
  113. кушать
  114. машина
  115. имя
  116. новый
  117. думать
  118. терять / глагол утраты объекта
  119. позволять / глагол соц. разрешения и запрещения
  120. полный
  121. близкий / приблизительный
  122. быстрый
  123. знать точно / извещать = информировать
  124. сухой
  125. значить
  126. simsa
  127. sampu
  128. ciska
  129. разновидность (сорт, вид, тип)
  130. верх / поверхность
  131. изменять (+Покупки(Обмен/Валюта)
  132. годиться
  133. время
  134. pinxe
  135. pilno
  136. одежда
  137. жить
  138. путь
  139. valsi

Free lunch coming.

'Lojban for Beginners' will meet you under a new name.

In better taste.

In less complexity.

Stay tuned!

Free lunch coming.

'The Crash Course in Lojban'. A classical work under a fresh name.

'Lojban for Beginners' redesigned to meet the new demands from the world community.

A better quality, a slimmer design.

Lojban as it is in year 2015.

Got your freebie pdf yet? Just click it.

http://mw.lojban.org/index.php?title=L17-01

The original idea by Robin Turner.

Initially revised by Nick Nicholas, the first fluent speaker of Lojban on the planet.

Reworked by La Gleki with the help of Lojbanic community throughout year 2014.

Just one click away!

http://mw.lojban.org/index.php?title=L17-01

Let the magic start here!

About the book

This textbook is or will be further inspired by:

Lojban textbook in Wikibooks, BASIC English, gua\spi, Pandunia lessons, toki pona with it's textbook, Sesperanto, Conceptual Semantic Categories of Lojban gismu, Viku lessons, Loglan3 textbook, Glosa textbook, Simple-E from Lognet'99, survival lingvo, Easy Loglan (Introduction and Description),N-paradigm, Standard Average European, Tok Pisin for beginners, Conlanger's Thesaurus, Introduction to Ceqli adapted for Lojban.

poi problems

Another method is to replace lo with zo'e and attach any number of relative clauses connecting them with je:

mi klama zo'e poi gusta lo kisto je poi berti lo tcadu
I go to something that is a restaurant of Pakistani cuisine and that is north of town.

You can use da/de/di if it is not used otherwise in the sentence (remember that the grammar of da and zo'e differs):

mi klama da noi gusta lo kisto je poi berti lo tcadu
I go to something which is a restaurant of Pakistani cuisine and that is north of town.

ke'a

We used ke'a in

do pu tavla lo be mi mensi poi ke'a nelci nai la .rikis.martin. vau lo dinske
You talked to my sister who doesn't like Ricky Martin about economics.

This ke'a replaces lo be mi mensi in the relative clause.

So ke'a is used in order not to repeat lo be mi mensi twice. We can't even use ri here instead of ke'a since ri refers to the last finished noun and then noun lo be mi mensi is not finished yet, since the relative clause is a part of the noun to which it is attached.

ra would be okay but it isn't particularly precise.

po and po'e

pe and ne are used as loose association only, like saying lo stizu pe mimy chair about a chair which you sit on. It's not really yours, but has something to do with you. There is just a relationshop between the two nouns.

stizu = x1 is a chair

If you want to say something is specific to that noun you can use po instead of pe.

lo gerku pe la alis can be applied to a street dog that Alice likes to play with.

But lo gerku po la alis can be applied only to a dog that, for example, was bought for her by her parents.

In many cases people prefer to use pe which is more vague rather than po.


Some other examples:

lo cukta po mi
My book
lo cipni po la .meilis.
Mei Li's bird
la .kokakolys. po do
Your Coca-Cola

Lastly, po'e is used for innate, intrinsic connections.

lo birka be mi or lo birka po'e mi = my arm
lo mamta be mi or lo mamta po'e mi = my mother
lo gugde po'e mi = my home country (a more precise analog is lo gugde poi mi pu jbena bu'u ke'a = country where .i was born)

As you can see, in many cases the second place of verb makes po'e unnecessary.

Because of all of that pe is used very frequent compared to po and po'e.

Resume. We have:

pe — restrictive relative clause. "which is associated with..."
ne — non-restrictive relative clause. "which is associated with..."
po — restrictive possessive relative clause. "which is specific to..."
po'e — restrictive innate relative clause. "which inherent to..."

L17-02

ti o ta

  • ti .o ta = this if and only if that (both or none)
    • ti na .o nai ta = this if and only if that (both or none) (the same meaning as .o)
  • ti .u ta = this whether or not that (this, and perhaps that)
    • ti .u nai ta = this whether or not that (the same meaning as .u)
  • ti .o nai ta = either this or that
    • ti na .o ta = either this or that (the same meaning as .o nai)

more about be and relative clauses

How would we say You talked to my sister — the one who doesn't like Ricky Martin — about economics? Let's take it by steps:

do pu tavla lo be mi mensi lo dinske
You talked to my sister about economics.
lo be mi mensi cu nelci nai la .rikis.martin.
My sister does not like Ricky Martin.
do pu tavla lo be mi mensi poi ke'a nelci nai la .rikis.martin. vau lo dinske
You talked to my sister who doesn't like Ricky Martin about economics.

Notice that it is possible to move be mi to the left of the verb mensi. This allows to attach poi to lo be mi mensi. In lo mensi be mi poi... the word poi will attach to mi: the sister of me who (I) ....

Also notice that we need to close the relative clause with vau so that lo dinske belongs to the main verb of our clause: tavla, not the verb nelci inside the relative clause. Otherwise, lo dinske would be a noun of nelci and not tavla — which is not really what you want.

Here's another example:

mi klama lo gusta be lo kisto
I go to the Pakistani restaurant.
lo gusta be lo kisto cu berti lo tcadu
The Pakistani restaurant is north of town.

Here we can't attach poi berti lo tcadu (that is to the north of the city) to lo gusta as lo kisto doesn't allow us to do that. We also can't move lo kisto to the left as lo kisto gusta would create a a compound verb (tanru). In this case we just repeat poi twice so that the second poi attaches to lo gusta.

To attach two or more relative clauses use je (and) to connect them, since both clauses are supposed to be true. For example,

lo be mi mensi poi na'e nelci la .rikis.martin. je poi do viska ke'a ca lo prulamcte
My sister who doesn't like Ricky Martin and whom you saw last night.
do bajra du'i lo mlatu
You run like a cat.
You run equally to the cat running [literally]
du'i — preposition. "equally to …" (comparing event to event)

As an event follows du'i in this example du'i is applied to the whole event of the clause. If the preposition is based on a verb whose x2 is an event then the preposition is applied to the whole clause where it resides:

du'i is based on dunlito be equal to. We can compare only objects with objects and events with events. Since an event is attached to du'i it is appliedю

selmaho

Particles (cmavo) are short words like lo, cu, je, pu, ca, ba, mi, do, ti. They can be divided by their meaning and function into groups that are called selma'o. E.g. pu, ca, ba belong to the group PU. This is just an illustration, there is no need in memorizing the names of selma'o. But it's convenient to memorize them by those groups.

klama

Well, here is the full definition of klama:

klama = x1 comes/goes to destination x2 from origin x3 via route x4 using means/vehicle x5

So

lo klama = a goer
lo se klama = a destionation gone to
lo te klama = an origin, starting point gone from
lo ve klama = a route
lo xe klama = a vehicle

Time of day

<--Let's imagine, though, that the time is not ten past eleven, but ten to eleven. We can say li pa no pi'e pano (10:50), but we can also say li pa pa pi'e ni'u pano, where ni'u is the Lojban minus sign (for negative numbers, not for subtraction) — what we are saying is 11:−10.

For half past eleven we can also use pi and say li pa pa pimu 11.5. I don't particularly like this method, but it is perfectly good Lojban. If we are using numbers for times, it is normal to use the 24-hour system, so 6 PM is li pa bi (18:00).-->

tails

23:32 < durka42> gerna: broda gi'e ca da brode 23:32 < gerna> not grammatical: broda gi'e ca _da_ ⚠ brode 23:33 < zahlman> what were you trying to express? 23:33 < durka42> ko'a ge broda gi ca da brode 23:35 < durka42> I thought the tag shouldn't interact with the connective

                unless I used {gi'ecabo} which I didn't                        

23:38 < zahlman> gerna mi ca da broda 23:38 < gerna> (0[{mi <ca da>} CU {broda VAU}])0 23:38 < zahlman> notice that {ca da} is not part of the selbri here 23:38 < zahlman> gi'e connects bridi-tails, so they have to start with selbri.

member vs. full set

10:04 < latro`a> gleki: I think in the gismu themselves, it is best if these are polymorphic
10:04 < latro`a> then you can specify later
10:05 < latro`a> unfortunately this does lead to odd interaction with quantifiers, because polymorphic selbri really are polyvalent
10:05 < latro`a> so {mi po'o se bangu la lojban} makes sense for one of the manifestations of {bangu}
10:06 < latro`a> in haskell this problem is solved by the typechecker: a polymorphic function comes with a dictionary that maps values of the variable input types to implementations
10:07 < latro`a> for example, if I define "f x = 3*x+1", I get a function that carries around a dictionary that tells it to use (*) and (+) from Int if x is an Int, from Double if x is a Double, etc.
10:08 < latro`a> it'd be interesting if we could somehow implement something like this
10:08 < latro`a> but we probably can't, which means you really shouldn't use quantifiers in polymorphic places
10:16 < latro`a> I suppose to a limited extent this could be done with SE, though this would get rather cumbersome if we had more distinctions than distributive/nondistributive and exhaustive/nonexhaustive
10:17 < latro`a> or NAhE
10:24 < gleki> latro`a: what should i use {poi se gunma} and {poi cmima}?
10:24 < latro`a> that's not enough, unfortunately, you have to tweak the selbri instead
10:25 < latro`a> because this isn't really about the type of the argument
10:25 < latro`a> skina with skina4=nonexhaustive and skina with skina4=exhaustive are different predicates
10:25 < gleki> to tweak? how?
10:26 < gleki> for skina i can say {mi catlu lo skina}
10:26 < latro`a> NAhE or SE are the syntax you want
10:26 < gleki> but about the other cases?
10:26 < latro`a> I'm imagining a NAhE for "x1 is exhaustive"
10:26 < latro`a> then "x2 is exhaustive" is {se NAhE* se}
10:26 < gleki> "this is a language of French people" = ko'a bangu lo fraso
10:27 < gleki> "this is a language of some French people (my friends)" = ko'a bangu lo fraso
10:27 < gleki> how to distinguish them?
10:30 < gleki> seriously i understood nothin from ur proposal
10:33 < latro`a> ko'a se NAhE* se bangu mi == ko'a is a language of me exclusively
10:33 < latro`a> where NAhE* is a new NAhE
10:33 < latro`a> this is a bit stronger than the {po'o} version
10:44 < Ilmen> coi. What do you mean by «exhaustive»?
10:48 < gleki> cipra: ko'a se na'e se bangu mi
10:48 < cipra> (ko'a [CU {se <na'e (¹se bangu¹)>} {mi VAU}])
10:53 < gleki> that would be nice ofc. to have them polymorphic but then what would be the difference between cmima and se gunma? ;)
10:59 < latro`a> cmima isn't really a thing in post-set jbo
10:59 < gleki> what is it then?
10:59 < latro`a> cmima?
10:59 < gleki> yes
10:59 < latro`a> it's the way you refer to sets
10:59 < latro`a> which we don't use
11:00 < gleki> so now we dont need {cmima} gismu at all?
11:00 < latro`a> not really no
11:00 < latro`a> se gunma is mass membership
11:00 < gleki> So "I am a member of this committee" would be ...
11:00 < gleki> en: kamni
11:00 < mensi> kamni = x1 (mass) is a committee with task/purpose x2 of body x3. |>>> Board of directors/trustees/cabinet (= trukamni, gritrukamni). See also bende, kagni. |>>> noralujv
11:01 < gleki> would be what?
11:02 < latro`a> mi se gunma ti noi kamni
11:02 < Ilmen> So, what is an exhaustive tersu'i? Do you mean things like gunma2 which doesn't require to list all the members of the mass?
11:02 < latro`a> being literal
11:02 < latro`a> maybe this will help
11:02 < gleki> latro`a: do you think i should remove {cmima} from my analog of L4B ?
11:02 < latro`a> mi se gunma mi joi do
11:02 < latro`a> mi na se NAhE* se gunma mi joi do
11:03 < latro`a> mi jo'u do se NAhE* se gunma mi joi do
11:03 < latro`a> this also gives you a way to extract the individual-blob from a mass, provided that makes sense
11:03 < gleki> why {na} in the first sentence?
11:03 < latro`a> {lo se NAhE* se gunma be ...} is unambiguously all of the individuals inside, in one lo-blob
11:04 < gleki> could u at least give translations?
11:04 < latro`a> {mi se gunma mi joi do} = I am a member of the mass {mi joi do}
11:04 < Ilmen> What means {lo ro se gunma be ko'a}?
11:07 < latro`a> {mi na NAhE* se gunma mi joi do} = I do not satisfy "x1 (individuals) is the collection of all individual members of x2"
11:05 < latro`a> it is a slightly more vague version of the same thing
11:05 < latro`a> "all of the members of ko'a that we are talking about"
11:05 < latro`a> but we might not be talking about *all* of them
11:05 < latro`a> this NAhE* is sort of an explicit UD-twiddler
11:06 < gleki> this is one of the three questions that stop me from completing a new gimste with examples, te sumti interactions, variable types, blackjack and hookers
11:06 < latro`a> {lo ro ve skina} may not contain the entire audience of the film, because we might just be talking about a small group of friends that watch movies together
11:07 < Ilmen> To me, skina4 is the *intended* audience, not the people who actually watch skina1
11:07 < Ilmen> ie pei
11:07 < gleki> as for "intended" te sumti this is actually my second question :D
11:07 < latro`a> to me that's the exhaustive skina4, but I regularly use ve skina for "watch"
11:08 < latro`a> I do not really like kinzga

necessary and sufficient

13:38 < Ilmen> does banzunu = se sarcu?
13:41 < latro`a> this is muddled by classical logic shenanigans
13:41 < latro`a> in classical logic, p => q (that is, ~p v q) is the same as p |- q
13:41 < latro`a> the latter is "one can infer q from p"
13:42 < latro`a> this is the definition of |-, more or less
13:43 < latro`a> in general necessary/sufficient is about |-, not =>
13:43 < latro`a> and |- is different in nonclassical logics, and is also different in human discourse
13:44 < Ilmen> So in this paradigm sufficiency is janai?
13:45 < latro`a> in classical logic, "p is sufficient for q" is "p => q", which is to say "p na ja q"
13:45 < latro`a> while "p is necessary for q" is "p <= q" which is to say "p ja nai q"
13:45 < Ilmen> What about the lojban "sarcu", should it be a mere logical connective, or should it express causation?
13:46 < latro`a> in human discourse, |- is more complicated than any simple logical operator
13:46 < latro`a> because we do not say "if the sky is red then mars is blue"
13:46 < latro`a> even though under the classical definitions this is true
13:46 < Ilmen> I'm not very familiar with |-. It's a metalogical operator?
13:46 < latro`a> |- is inference
13:47 < latro`a> xatra: tell Ilmen the name for the symbol |- in english is "turnstile"
13:47 < latro`a> bah
13:48 < latro`a> Ilmen: the name for the symbol |- in english is "turnstile"
13:48 < Ilmen> got disconnected .oi
13:48 < Ilmen> did I miss something?
13:48 < latro`a> Ilmen: the name for the symbol |- in english is "turnstile"
13:51 < Ilmen> je'e
13:51 < Ilmen> So, what {sarcu} means?
13:52 < latro`a> there's causality there, definitely
13:52 < latro`a> but it's event-causality
13:53 < Ilmen> nibli is more a sufficient condition or a necessary conditiont?
13:53 < latro`a> nibli is "x1 is logically sufficient for x2"
13:53 < Ilmen> ba'a ie
13:54 < Ilmen> {banzunu} is the sufficient version of sarcu
13:55 < Ilmen> We lack a necessary-nibli
13:55 < Ilmen> ja'o
13:59 < Ilmen> X sarcu Y = lonu X tolfau cu banzunu lonu Y tolfau  (if I'm not mistaken)

take any apple (not more than one)

14:01 < menli> {lo ro plise goi ko'a zo'u ge ro da poi me ko'a zo'u do zifre lo ka lebna da gi ko'oi do lebna pa ko'a} -- what's wrong with this?

Lojban doesn't force you to use any particular philosophy, religion or political beliefs.

Explain nai

Unlike when used after verbs and pronouns nai after interjections means not just "no" but the opposite attitude.

But explain after BAI

Variable types

12:32 <@xalbo> Some object/event places are events with possible raising, but some (what I've called Objects of Cognition) really aren't. Although those hold a lot more than just object/event. 12:34 < zahlman> "objects of cognition"? 12:38 < menli> For example, zmadu-1&2, drata-1&2, melbi-1, simsa-1&2 have no requirement on the sumti type 12:42 <@xalbo> pensi2 is the prototype OoC, for me. Similarly, melbi1. "Anything the human mind can conceive of and opine about." 12:43 <@xalbo> {zmadu} is a more complicated thing, since zmadu1 and zmadu2 do have some more restrictions (namely, that they're both compatible with the open place in the zmadu3)

spisa - pagbu

  • Ilmen: piece of wood, portion of water, chunk of whatever. They are a part of a substance/mass noun/uncountable goo
  • lump/cluster?
  • So when you use spisa you attribute substanceness to something, and spisa1 is a chunk of it. Whereas pagbu is a part-whole relation
  • My arm is a part of my body
  • かたまり (n) lump; mass; bundle; clump; clod; cluster; (P)
  • pagbu is partially something while spisa is part of something?
  • so some driftwood I found on the beach is a piece of wood (spisa) but if I break it in half I then have two pagbu. is that right?
  • pagbu = a part of something (not necessarily delimited)

da

14:07 <@xalbo> I think the point isn't that {ci da plise} is talking about three apples, it's saying that the number of apples is three. "How many apples are there?" "Three". 14:07 < mukti> So not to derail the goatlegging, but earlier I asked about whether {ro} performs existential import. The consensus seemed to be that it does not. Does this mean "The problem of 'any'" section in CLL needs an update? {ro da poi klama le zarci cu cadzu le foldi} is read to say "in fact, there are people who go to the store", and more to the point: "Lojban universal claims always imply the corresponding existential claims as wel 14:07 < mukti> l" 14:07 < selpahi> The thing is that {da} does not really make reference. If you pick out three from among the five which are all in the same place, it is pretty hard to argue that the other two are not in the domain. So I'd suggest {le} or {lo} here 14:07 < ldlework> selpahi: I think the problem is that I define the UoC with my speech 14:07 < ldlework> The state of the world, objectively, doesn't. 14:08 <@xalbo> mukti: Huh, I hadn't realized that CLL came down firmly on the side of existential import like that. Very interesting. 14:08 < selpahi> mukti: Yes, this is a known problem 14:08 < mukti> xalbo: I feel very silly asking the kinds of questions that I am today. Yeah. 14:08 < mukti> selpahi: Good to know.

06:43 < mukti> "I want to marry any rich woman" is different from either the

              specific or non-specific reading of "I want to marry a rich
              woman"

06:44 < gleki> is the second da poi ricfu ninmu cu jai se djica mi fai lo ka mi

              se speni

06:44 < gleki> ? 06:47 < gleki> then the first is {ro ricfu ninmu ka'e jai se djica .. 06:48 < menli> {mi djica lo nu mi speni pa da poi ricfu gi'e ninmu} vs {pa da poi ricfu gi'e ninmu zo'u mi djica lo nu mi speni da}

Fixed issues

  • fxd?The modern interpretation of ka and of other abstractors.
  • fxd?That if you want a Lojban name, it can be a selbri
  • fxd restore {le} explaining what it means? (we have {le} retaining in old l4b tables and unexplained)
  • fxd When you cover the vowels, put .ybu in the table. It is confusing otherwise and will be more useful that way for quick reference. You can still explain the vowels as you currently do after the table.
  • fxd? cmavo can be written without spaces
  • fxd? That vi does not mean "at location ..." (that's bu'u), but rather that it tags a spatial distance, like zi tags a temporal one.
  • fxd? me SUMTI moi
  • fxd? Then, I would also consider renaming the chapter a bit. Many concepts mentioned do not, technically, fall under "grammar" (such as pronunciation).
  • fxd? Also, I maybe wouldn't call it "newbies". I'm a computational scientist myself, so I understand where the term comes from, but it is not something that has really passed into the language. I never hear somebody calling somebody else on the street a "newbie" (the "for dummies" actually uses a more recognized word).
  • fxd? Other than that, it seems well structured, if a bit short at some points. Remember that concepts which probably seem natural and easy to you have never been seen before by your readers.
  • fxd? This is a problem I noticed with wave lessons continued: Some of the things mentioned (not many, thankfully), require quite a bit of thinking and research in figuring out, because the authors assumed the readers would just understand it right away.
  • fxd? When introducing zo «ti» and zo «tu», why not fill in the blank with zo «ta» and give them their first series? (You can also appeal to Spanish for differentiating betwixt these words, since English does not recognize such a distinction but many English speakers are familiar with Spanish, which does.)
  • fxd? When mentioning emotions in the introduction, I presume that you are referring to attitudinals? If so, are they completely novel to human language? Also, why not throw in reference to evidentials, discursives, vocatives, and erasure? Those are all sort of bundled together in my mind because they are rather simple and stand-alone in nature, but pack a mighty punch.
  • fxd? The introduction keeps beginning paragraphs with "Lojban". I am not sure whether it is stylistic or not. If so, leave it as it is if you want (although I would check to see that you have a prime number of "Lojban"'s at the beginning of new paragraphs- humans find such combinations aesthetically pleasing). If it was not intentional, you may consider revising the wording a bit. It can cone across as a bit clunky. (I know that this is a draft, but I thought that we might as well address it now.)
  • fxd I would give some mention to tenses and connectives, since Lojban is rather careful and unique in how it implements them in their totalities.
  • fxd Also, are you going to gloss translations, simply give colloquial English translations, or both? If the last, will both the gloss and the colloquial translation be italicized?

L17-05 - NO FIX issues

Indefinite articles

In English we say "a table" and "an egg". The "a" and "an" are indefinite articles. Lojban does not use indefinite articles. It is enough to say lo sovda for egg or "an" egg; lo jubme for table or "a" table, and so on, when referring to "some unspecified" egg or table, or "any" egg or table. The definite article "the", used to specify a particular object being discussed, is replacing lo to le lo fetsi - a female, le fetsi - the female or just she (a specific female, the one we are discussing).

Other vocatives

co'o is the farewell word, corresponding to Goodbye, Farewell. Lojbanists signing off on e-mail often end with something like co'o mi'e .bob. — this is equivalent to putting your name at the end of your email in English as a signature, and translates as Goodbye; I'm Bob.

  • Here are some more rare vocatives:
  • co'oi is the greeting/parting word much like Italian ciao: it corresponds to Hello / Bye.
  • ju'iHey!, with which you draw someone's attention, and
  • fi'iWelcome! At your service!, with which you offer hospitality or a service. (It's what you say to a visitor; you wouldn't say it over the phone, for instance, unless your addressee is calling from the airport and is on their way over).
  • ki'eThank you and the appropriate response is not fi'i (You're welcome doesn't mean you're being visited by some guests), but the simple acknowledgement je'e.
  • je'e corresponds to Roger! in radio-speak, and right or uh-uh in normal English: it confirms that you've received a message. If you haven't, you say je'enai instead (of course); in normal English, that would be Beg your pardon? or Huh?
  • In case you haven't received the message clearly, you can explicitly ask for the speaker to repeat whatever they said with ke'o.
  • Similarly, be'e signals a request to send a message (Hello? Are you there?), and re'i indicates that you are ready to receive a message. It's what you say when you pick up the phone — which in English also happens to be Hello?, but in Italian is Pronto or Ready!
  • mu'o is what you say when you explicitly make it another speaker's turn to speak: it's the Over! of radio.
  • vi'o acknowledges a request, and promises to carry it out: in radio talk this is Wilco!, and in normal English OK or All right, I will (or for that matter, Consider it done!)

Vocatives take nouns after them. However, the rule is that you can drop lo making it more vague:

coi gleki
Hello, friends!
can mean both
coi lo glekiHello, a happy one or
coi la glekiHello, Happy (a personal name) depending on context.

  • "12:39 < phma> ra'o can also modify mo or du, neither of which makes much sense, or no'a

12:40 < gocti> va'o su'o cipra gerna cu cmavo ma'oi ui :p 12:42 < phma> ""do mo ra'o? — tavla mi"" is equivalent to ""do mo? — do tavla mi"" 12:42 < phma> and here's my example of no'a ra'o: 12:43 < gocti> ua 12:43 < phma> la .djan. klama le zdani be vo'a pu le nu la .meris. no'a 12:43 < phma> John went to John's house before Mary did 12:44 < phma> la .djan. klama le zdani be vo'a pu le nu la .meris. no'a ra'o 12:44 < phma> John went home before Mary did 12:44 < phma> In the second sentence Mary went to Mary's house. In the first, she went to John's."

  • Quotes in e-mails frequently start with do cusku di'e (di'e is a particle that means ‘the following’) as the Lojban equivalent of You wrote.
  • Tip: lu … li'u is intended to quote grammatical pieces of Lojban — ideally, entire sentences, rather than individual words. For smaller chunks of Lojban, which do not necessary make sense in isolation, the proper quotation words are instead lo'u… le'u, the ‘error quotes’. For example, ro lo mi pendo cu klama makes sense in Lojban as a sentence, and can be enclosed in lu … li'u. But if you want to say what goes before pendo in the sentence, ro lo mi does not make that much sense on its own. So you would quote that sentence fragment, not as lu ro lo mi li'u, but as lo'u ro lo mi le'u.

  • You can divide up the continuum even more finely. If you want to say that you have an only weak attitude, you can add the particle ru'e to the corresponding interjection. sai is used for expressing strong degree of an attitude. Extremely strong attitude is cai. This gives you a seven-part scale:
    cai > sai > (nothing) > ru'e > cu'i > nai ru'e > nai > nai sai > nai cai
    • If you want to say That is really gross!, you'd say .a'unaisai. And if you want to say Oh my God, that is the most interesting thing in the world since the very invention of Lojban! then you would say .a'ucai. cai is used very seldom as strong emotions happen rarely.
  • lowering. kantu) is assumed.
  • ti sarni lo stero be li pai fe'i re lo mitre This triangle has an angle of π/2 radians facing the side of one meter in length.

place here issues for which CC can function without solving them:

  • re'o - splicing?
  • cinza - another word for chopsticks needed?
  • ??? x2 is danti, x3 is propulsion
  • off definition says fenso2 must be complete specification
  • fatci. mucti are not real things
  • https://ru.wikipedia.org/wiki/Sensu_stricto - do'anai a bit similar
  • erotic vs. {seksi}, a brivla needed
  • restore xrani4
  • is funsi2=fancu2 ?
  • rai - add ne
  • is ke'e needed?
  • is carna clockwise solved?
  • {zu} is modifier, not just a tag. It interacts with the previous tag. Compare {puzu} and {pudo'ezu}. The latter is similar to bare {zu}.
  • In natural languages, reflexives almost always refer back to subjects; and in Lojban, the x1 place is as close as you will get to a subject. The difference is, when you have this kind of embedding, the reflexive can refer back to the subject of the verb it is immediately tied to (short-distance reflexive like our lo nei), or it can refer all the way back to the subject of the entire sentence (long-distance reflexivelike our vo'a). Now, herself in English is a short-distance reflexive: if Jasmine knows that Alice loves herself, then Jasmine knows that Alice loves Alice, not Jasmine. In some languages reflexives are short-distance. In other languages like Chinese reflexives can be either long-distance or short-distance. Or languages may have long-distance reflexives distinct from short-distance just like in Lojban.
  • Native Americans generally translate their name when speaking English, partly because they have meaningful names, and partly because they don't expect the wasichu to be able to pronounce words in Lakota, Cherokee or whatever!
de'i li pi'e bi pi'e ro mi na gunka
Every August I don't work.

Here, ro refers to the year

  • The default calendar is the standard Western one as it is international. If you want to use, for example, the Arabic or Chinese calendars, you can put lo xrabo or lo jungo in the fifth place. (As always, context is important — in a discussion of Islamic history we would probably assume that the Arabic calendar was being used).
other date related

And here are the names of months of Gregorian calendar: <tab class=wikitable head=top>

English name Lojban name meaning January lo pagmese 1st month February lo regmese 2nd month March lo cigmese 3rd month April lo vogmese 4th month May lo mugmese 5th month June lo xagmese 6th month July lo zegmese 7th month August lo bigmese 8th month September lo sogmese 9th month October lo daugmese 10th month November lo feigmese 11th month December lo gaigmese 12th month </tab>

It is easier to remember them if you notice that they the first letters remind of numbers. Thus, January is the first month and start with pa (1) and so on.

Calendars in other cultures

The names of days of the week and months match international standards. However, there can be a need to describe conventions for cultures which for example do not use a seven-day week. In ordinary speech you are free to create compound verbs for that. For example, you can call the first month of classical Chinese calendar as lo jungo pavma'i (literally Chinese January although the first month in the original Chinese calendar rather starts in February). And lo jungo pavdei could be the first day of the Chinese ten-day week. (jungo means x1 is Chinese).

The same logic can be applied if the seasons where you live don't match this pattern. For example, the rainy season or monsoon could be lo carvi citsi (from carvi = rain, and citsi = season). Here are some I made up for fun to give a better idea of the weather in Britain:

lo lenku carvi citsi the cold rain — spring
lo mligla carvi citsi the warm rain — summer
lo brife carvi citsi the windy rain — autumn
lo dunja carvi citsi the freezing rain — winter

Later in this course we'll see how to create new words in the form of lujvo and zi'evla and those words will have precise meanings and thus become terms. You'll actually need a pretty good knowledge of Lojban to make up lujvo on the spot, but we'll learn how to make some simple lujvo later on in this course.

aspects

We can also contrast do ca ze'i pu klama with do pu ze'i ca klama. The first event of traveling has one endpoint in the present and extends a little towards the past, while the second event has one endpoint in the past and extends only to the present (that is, slighty into the past or future) of that endpoint.


Note that mi ba tavla is similar to mi pu'o tavla, and likewise with ba'o and pu. Why do they seem reversed in sounding? Because event contours view the present as seen from the point of the process, whereas the other tenses view events seen from the present.


All of these tenses have been describing stages of a process which takes some time (as shown on the graph above; those tenses above the event like). But many of the event contours describes point like stages in the process, like its beginning. As is true of ca and bu'u, they actually extend slightly into the past and future of that point, and need not to be precise.


For this kind of aspect, English normally just uses verbs: start, stop. Lojban likewise allows you to use distinct verbs to express these notions: cfari, mulno, and sisti. Using aspects just lets you express things more succinctly; and with Lojban the way it is, anything that makes things more succinct comes in handy.

Most of the time, though, processes actually end at their natural ending; this is what makes it natural. Trains are not usually late, and people usually retrain themselves to eat only edible food.

To convert de'a/di'a to space use fe'e.

All event contours in one diagram: contours.png

Vertical line signifies the time for the natural beginning and natural end of an event.

Horizontal lines show the event contours.

What would .o'onai ri'u nu lo prenu cu darxi lo gerku pu mean?

o'o: interjection: patience. The full scale is o'o (patience), — o'ocu'i (tolerance), o'onai (anger)
darxi = x1 beats/hits x2 with instrument x3 at locus x4

Answer: (Anger!) To the right (of something, probably me) and in the past (of some event), something is an event of a person beating a dog. or A man hit a dog to my right!

mi klama pu'o lo nu do citka
As you are eating, I'm about to go.

pu'o needs to be demonstrated by an example. What does .ui mi pu'o se zdani mean?

Answer: Yay, I'll begin to have a home.

jmive
x1 is alive by standard x2

What does .ui mi pu zi ze'u jmive mean?

Answer:

(I'm happy) I live from a little into the past and a long way towards the future or past (obviously the future, in this case) of that event.
I am young, and have most of my life ahead of me :-).

Just to underline the similarity with spacial tenses, let's have another example, this time with spacial tenses:

bu'u ve'u ca'u zdani
This home extending ahead is huge!
Extending a long space from here to my front is a home. [literally]
-

In Lojban, we also operate with an event's natural beginning and its natural end. The term natural is highly subjective in this sense, and the natural end refers to the point in the process where it should end. You can say about a late train, for instance, that its process of reaching you is now extending beyond its natural end. An undercooked, but served meal, similarly, is being eaten before the natural beginning of the process.

Precedence

We should be able from that to say

.i la .flufis. cu ractu .i je ro lo ractu ze'i jmive .i se ni'i bo la .flufis. ze'i jmive

Right? Actually, no we can't: bo has the function of connecting sentences through prepositions, because it connects sentences on its own. And when it does, it connects them tighter than .i je does. This means that .i se ni'i bo connects only to the immediately preceding sentence — not to the preceding sentence pair! So Fluffy's death is presented as a consequence of rabbits not living long — not a consequence of both rabbits not living long and Fluffy being a rabbit.

However, if we put the two clauses in a single sentence, then none of this is an issue: the conclusion will attach to both clauses, but will still attach to a single sentence:

.i ge la .flufis. cu ractu gi ro lo ractu ze'i jmive .i se ni'i bo la .flufis. ze'i jmive

There is also a forethought conjunction for compound verbs: these are gu'a, gu'e, gu'o, gu'u. And the second compound part is connected with gi. So if we want to say that Alice fancies men that are, if funny, then also handsome, the afterthought version is

la .alis. cu nelci ro lo melbi na ja xajmi nanmu

To make this slightly (but only slightly!) more comprehensible, we can put this in forethought mode:

la .alis. cu nelci ro lo gu'a nai melbi gi xajmi nanmu

There are no forethought versions of clause tail conjunctions. In practice, however, two clauses connected by ge can be clause tails just as easily as a full clause: there is no real distinction in meaning between the two.

Situation: "My garden"

ti me lo purdi pe mi .i mi cadzu bu'u py. This is my garden. I walk in the garden.
.i mi tirna lo cipni poi sanga .i lo sance be cy. na cladu I hear birds singing. The sound of birds is not loud.
.i lo sance cu tolycladu .i lo rilti cu pluka .i lo tonga cu galto'a The sounds is quite. The rhythm is nice. The tone is high.
.i tcima fa lo solri .i mi catlu lo tsani The weather is sunny. I look to the sky.
.i mi na viska lo solri .i lo dilnu cu fanta lo solri gusni I don't see the sun. Clouds covers its beams.
.i ku'i mi ca'o ganse lo glare gau lo solri But I can still feel the heat of the sun.
.i mi viska lo plise tricu .i ri clani .i lo plise cu crino I see apple trees. The trees are tall. The apples are green.
.i mi klama lo plise tricu .i mi ganse lo lenku ni'a lo tricu I go to the apple trees. I feel cold under the trees.
.i mi jdice lo ka sumne lo plise .i mi sumne lo panci I decide to smell an apple. I smell it's flavour.
.i lo panci cu pluka .i mi denzalvi lo plise The flavour is pleasant. I chew the apple.
.i mi smaka lo plise .i ri titla I taste the apple. It's sweet.
.i mi klama lo crane .i mi viska lo flora I go forward. I see flowers.
.i mi co'a zutse tezu'e lo ka sumne lo panci be lo flora I sit down to smell their flavour.
.i lo mlatu cu klama mi .i mi palpi lo mlatu A cat comes to me. I touch the cat.
.i lo sefta be ri cu ranti je xutla .i loi kerfa cu xutla Its surface is soft. loi kerfa is soft.
i mi co'a sanli .i lo purdi pe mi melbi I stand up. My garden is beautiful.
.i mi cinmo ma .i lo ka gleki i mi cinmi lo ka na badri What are my emotions? Happiness. I feel that I am not sad.


Task. Answer these questions (close the right part of the table):

lo sefta be lo mlatu cu mo xutla or ranti je xutlaWhat is the cat's surface? — Smooth, smooth and soft.
xu do cinmo lo ka badri je'unaiDo you feel sad? — No.

A similar set of examples with ga'a:

do ga'a mi fenki
To me (from my observation) you are crazy.
mi ne se ga'a do tirna
I'm listening to you.
I (to observe you) hear. [literally]
ga'a = preposition from zgana: to observer ...

The last sentence is similar to:

mi zgana do lo ka tirna
I observe you by hearing.

Types of nouns: nouns from verbs, la-names and pronouns

There are three types of nouns in Lojban:

  1. nouns started with lo
  2. nouns started with a number, then lo, then a verb word
  1. nouns started with a number, then a verb word
  2. name is la + a verb.
  3. pronouns are miI, tathat and others

So as they mostly work the same way we in Lojban there is one word covering all of them - sumti.

Terminology in clauses

Let's describe the structure of Lojban clause (bridi). The main verb, or predicate (se bridi or selbri in Lojban) describes relationships of nouns. It can be represented as a single verb word (selbrivla) or as a compound verb (tanru).

Here are some examples of nouns and main verbs.

ti ladru
This is milk.

Here ti is a noun and ladru is the main verb consisting of one verb word.

lo mlatu cu sutra pinxe
A cat quickly drinks.

Here lo mlatu is a noun (sumti) and the compound verb sutra pinxe works as the main verb (selbri).

Also you can add prepositions (sumtcita) like ca:

le prenu ca citka
The person now eats.

So in other words.

bridi = optionally one or more sumti + one selbri + one or more sumtcita

or in English

clause = optionally one or more nouns + one main verb of the clause + one or more prepositions.

tanru, or compound verbs consist of two or more verb words. Each left verb word is called seltau compared to the right one called tertau.

Morphology of verbs

Verbs (selbrivla) are divided into 4 groups by their form:

  1. gismu, or root-words are main building blocks of Lojban vocabulary. gismu are easy to recognise, because they always have five letters, in the form
    CVCCV — for example, ladru, gismu, sumti, or
    CCVCV — for example, mlatu, cmene, bridi, klama
    where C=consonant and V=vowel.
    Verbs in the following forms are created when there is no appropriate verb in gismu list:
  2. lujvo, or compound words. They are created from short building blocks (called rafsi) used for mnemonic purposes.
    Examples are: retsku, kargau
  3. zi'evla, or free words. They are usually created for specific concepts and things like igloo (iglu in Lojban), spaghetti (spageti in Lojban).
  4. cmevla, or name words.

Task. Close the right part of the table. Which of the following Lojban words are selbrivla, cmevla (remember, they always end in a consonant), neither?

Note: the full stops are removed in the cmevla below to make the task a bit more tricky.

lojban cmevla
karce selbrivla
robin cmevla
mi cmavo
mlatu selbrivla
cukta selbrivla
fa'a cmavo
to'o cmavo
ian cmevla
ba cmavo
spageti selbrivla

lo

Notice that ro lo means each, every (whereas ro means all).

Finally, when defining cats with lo it's possible to specify how many of them exist: Template:Lo re mlatu cu pinxe lo ladru

me'i ci lo mu mlatu cu pinxe
Less than three out of five cats are drinking.

Yes, lo prenu can mean both one person or people (in plural) depending on context.

Lojban can be very vague. As one adds more words, the meaning becomes more precise.

Finally, we can put a number after lo:

lo mu plise cu kukte
Apples are tasty. And they are five in number.
lo pa no prenu cu citka
People (who are 10 in number) eat.